Large language model agent as a mechanical designer

Агент на основе большой языковой модели как конструктор-механик
Amir Barati Farimani, Yayati Jadhav
2026-02-11

2D truss structuresLLM-guided optimizationfinite element methodlarge language model agentsmechanical design
Traditional mechanical design relies on iterative refinement cycles driven by expert heuristics and computationally expensive Finite Element Method (FEM) analyses to meet performance targets. While machine learning approaches have been developed to automate portions of this process, they typically demand large datasets, substantial computational resources, and remain narrowly tailored to specific domains, thereby limiting their generalizability across diverse design problems. We propose a framework coupling a pretrained Large Language Model (LLM) with an FEM module to autonomously generate, evaluate, and refine structural designs based on performance specifications and quantitative feedback. Without domain-specific fine-tuning, the LLM leverages general reasoning to iteratively propose candidate structures, interpret simulation results, and implement mechanically informed modifications. Demonstrated on 2D truss structures, the approach navigates complex design spaces, balances competing objectives, and identifies appropriate stopping points when optimization yields diminishing returns. Compared to traditional methods like NSGA-II, LLM-guided optimization achieves faster convergence with fewer FEM evaluations in highly discrete, multi-faceted design spaces involving dynamic node generation and discrete member sizing. Experiments across multiple temperature settings and model scales reveal that smaller models achieve slightly better constraint satisfaction with fewer iterations, while lower temperatures produce more consistent performance. These findings position LLM agents as a promising new class of natural language-based, reasoning-driven optimizers for engineering design.
1
A framework couples a pretrained large language model with finite element analysis to autonomously generate, evaluate, and refine structural designs without domain-specific fine-tuning.
2
Compared with NSGA-II, LLM-guided optimization converges faster and requires fewer FEM evaluations for highly discrete designs involving dynamic node generation and discrete member sizing.
3
Smaller models achieve slightly better constraint satisfaction with fewer iterations, while lower sampling temperatures yield more consistent performance across experiments.
4
The LLM interprets quantitative simulation feedback and applies mechanically informed modifications while balancing competing objectives in complex 2D truss design spaces.
5
The agent identifies stopping points when further optimization produces diminishing returns, supporting efficient iterative design.

2D truss structural designs

autonomous LLM-guided generation, FEM-based evaluation, and iterative refinement of structural designs under performance constraints

Publication Details
Publication Date
2026-02-11
Journal
Publisher
ISSN
Cited by
13
Access Type
Author Information
Authors
Amir Barati Farimani
Yayati Jadhav
Explore further
Open the scid.ai AI chat with a ready-made request: it will find papers on a similar topic and help build a literature review.
Find similar papers in the chat →
Make a presentation
100%