Reproduce “A Multidisciplinary Design and Optimization (MDO) Agent Driven by Large Language Models” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Bingkun Guo, Wentian Li, Xiaojian Liu, Jiaqi Luo, Zibin Yu, Dalong Dong, Shuyou Zhang, Yiming Zhang
The MDO Agent, powered by large language models, orchestrates an end-to-end semi-automated mechanical design process, moving from natural language intent to optimized engineering solutions. This system significantly streamlines design cycles, demonstrating over 70% automation in parametric modeling and 70-75% in FEA simulation.
Key insights
- Large Language Models can effectively bridge the semantic gap between high-level human design intent and formal, executable engineering representations.
- A multi-agent architecture combined with specialized AI techniques (RAG for concept generation, VLM for CAD correction, ReAct for tool orchestration) enables robust end-to-end automation of complex mechanical design processes.
- Autonomous invocation of external engineering tools through LLMs significantly lowers technical barriers and enhances accessibility to advanced simulation and optimization software.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.