Date of Award
5-2026
Document Type
Thesis
Degree Name
Master of Science (MS)
Department
Mathematics and Systems Engineering
First Advisor
Luis Daniel Otero
Second Advisor
Juan Camilo Avendano
Third Advisor
Juan Carlos Palacios
Fourth Advisor
Gnana Bhaskar Tenali
Abstract
Model-Based Systems Engineering (MBSE) is quickly becoming the go-to practice of choice for systems engineers in today's defense, aerospace, and safety-critical systems. Yet, the first step of manually translating requirements to formal SysML models is a manual and time-consuming process. While there has already been some automation in certain parts of the SysML model creation process such as Natural Language Processing (NLP)-based requirement classification, and human-in-the-loop methods, there is currently no open end-to-end solution that takes natural language textual requirements as input and outputs validated SysML-consistent models complete with diagrams and full traceability.
This thesis presents a framework that is aimed at addressing this gap by presenting a rule-based, human-in-the-loop framework that automates the upstream creation of SysML v1.7-aligned models from natural language requirements. This proposed pipeline framework combines NLP with a deterministic and transparent rule-based parser, human-in-the-loop integration, automated diagram creation, and export of standard file formats. This proof-of-concept pipeline was evaluated on a thermostat system case study to validate that the pipeline can successfully generate SysML1.7 aligned models as well as the standard files of the system’s elements and relationships. This work directly addresses a critical gap in the MBSE literature and sets the foundation for future work to expand upon and work towards the goal of fully automating the process from textual requirements to a full SysML model.
Recommended Citation
Gorenflo, Spenser Harlow, "Toward a Modular, Human-in-the-Loop Framework for Automated SysML-Aligned Model Generation from Natural-Language Requirements" (2026). Theses and Dissertations. 1712.
https://repository.fit.edu/etd/1712