Exploration of Language Models in Automating System Design Diagrams from Natural Language Specifications

Author

• Candice Chambers, Florida Institute of TechnologyFollow

Date of Award

5-2026

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Computer Engineering and Sciences

First Advisor

Siddhartha Bhattacharyya

Second Advisor

Thomas Eskridge

Third Advisor

Khaled Slhoub

Fourth Advisor

Ryan T. White

Abstract

System design is the fundamental process of defining the architecture, components, and interactions of a system to satisfy specified requirements. In modern engineering, stakeholders typically require system engineers to produce System Design Diagrams (SDDs) to clearly visualize a system's functional and non-functional requirements. By leveraging Model-Based Systems Engineering (MBSE) modeling languages, SDDs can be used to employ formal methods for verification and validation early in the system's life cycle, long before production begins. However, manually generating these SDDs remains a persistent challenge in practice, as the process is tedious, error-prone, and highly time-consuming, often requiring deep expertise in specialized modeling languages. To address these challenges, we propose SysGenAI, a novel framework that takes a non-traditional approach to automating SDD generation. SysGenAI is a Language Model (LM) driven framework that integrates a formal grammar, enabling it to automatically transform natural-language (NL) descriptions of any system into structured SDDs, specifically, Architecture Analysis and Design Language (AADL) and Systems Modeling Language (SysML) models. A key strength of SysGenAI is its transparency. Rather than silently producing incorrect or incomplete outputs, the framework explicitly notifies users when generated diagrams do not meet correctness criteria, achieving a 68\% success rate in SDD generation while maintaining output integrity. The SysGenAI framework represents a significant and promising step toward fully automating the generation of SDD diagrams from NL specifications, with broad implications for reducing engineering overhead and accelerating system design workflows.

Recommended Citation

Chambers, Candice, "Exploration of Language Models in Automating System Design Diagrams from Natural Language Specifications" (2026). Theses and Dissertations. 1640.
https://repository.fit.edu/etd/1640