Committee Chair

Wu, Weidong

Committee Member

Owino, Joseph; Liang, Yu; Fomunung, Ignatius; Onyango, Mbakisya A.

Department

Dept. of Civil and Chemical Engineering

College

College of Engineering and Computer Science

Publisher

University of Tennessee at Chattanooga

Place of Publication

Chattanooga (Tenn.)

Abstract

This research investigates the feasibility of fine-tuning a domain-specific vison-language and large-language for truss structural analysis. General-purpose AI models often struggle with engineering-specific problems due to insufficient domain knowledge. To address this, we propose a hybrid workflow for truss analysis via the stiffness method as a case study. The project leverages a curated dataset of 27 truss templates and expanded through geometric augmentation, load randomization, and support variations. Llama 3.2 Vision Instruct was fine-tuned using the parameter-efficient fine-tuning to produce truss description from images, and T5-large was fine-tuned to convert these text description into JSON format for analysis using the stiffness method. Model performance was evaluated against the ground truth for node coordinates, elements, loads, and support conditions. This research demonstrates the potential of fine-tuned domain-specific language models to automate engineering analysis and design workflows, offering engineers and students a practical tool for rapid and accurate structural analysis.

Acknowledgments

I would like to express my sincere gratitude and thanks to Almighty God for His blessings and guidance. I am grateful for the continuous support, encouragement, and guidance provided by my advisor, Dr. Weidong Wu, from my first semester. I also thank all the members of my committee for their valuable feedback. Dr. Onyango and Dr. Fomunung provided thoughtful feedback on my work, and Dr. Liang helped me build a foundational understanding of machine learning, which was invaluable to this research. I thank my department and the program for the academic support and the computer facilities provided. In addition, I would like to thank the UTC Information Technology HPC Team for providing computing facilities and prompt support. Finally, I am grateful to my friends, including Mithu Chanda, Khowshik Dey, Nishat Tasnim Nisha, Kironmoy Paul Shourov, and many others; my parents, Rupam Dey and Priti Dey; and my wife, Anindita Chowdhury, for their constant support and motivation.

Degree

M. S.; A thesis submitted to the faculty of the University of Tennessee at Chattanooga in partial fulfillment of the requirements of the degree of Master of Science.

Date

5-2026

Subject

Machine learning; Natural language processing (Computer science); Structural analysis (Engineering)

Keyword

Truss structural analysis; large language models; vision-language models; domain-specific fine-tuning; stiffness method; structural engineering automation

Discipline

Computational Engineering

Document Type

Masters theses

DCMI Type

Text

Extent

xv, 74 leaves

Language

English

Rights

http://rightsstatements.org/vocab/InC/1.0/

License

http://creativecommons.org/licenses/by/4.0/

Date Available

5-31-2027

Download

Available for download on Monday, May 31, 2027

Share

COinS