Skip to main content Skip to navigation

Harnessing Multi-Agent LLMs for Complex Engineering Problem-Solving: A Framework for Senior Design Projects

Project Overview

The document examines the use of Multi-Agent Large Language Models (LLMs) in engineering education, specifically focusing on their role in enhancing complex problem-solving skills for senior design projects. It introduces a collaborative framework where multiple LLM agents, each embodying distinct expert perspectives, work together to address multidisciplinary challenges. This approach not only fosters critical thinking and ethical considerations but also emphasizes real-world problem-solving, ultimately aiming to improve educational outcomes and better equip students for future professional challenges. Findings indicate that Multi-Agent LLMs create a more inclusive and effective learning environment compared to traditional single-agent systems, suggesting their potential to transform educational practices in engineering and possibly other fields.

Key Applications

Multi-Agent LLMs for Senior Design Projects

Context: Educational context for engineering students undertaking complex, real-world senior design projects.

Implementation: Implemented through a framework where various LLM agents simulate expert perspectives and engage in collaborative dialogues to solve problems.

Outcomes: Enhanced critical thinking, collaborative skills, and a richer understanding of complex engineering challenges. Improved accuracy in project evaluations compared to single-agent systems.

Challenges: Need for effective prompt engineering and possible reliance on human input for optimal performance.

Implementation Barriers

Technical Barrier

Current LLMs require precise prompt engineering and may not fully simulate expert behavior without human oversight. Future work could focus on developing more autonomous multi-agent systems with better integration of external tools and advanced prompting techniques.

Proposed Solutions: Enhancing the autonomy of LLMs and improving their integration with external tools could reduce the need for precise prompt engineering.

Educational Barrier

Students may struggle with prompt design and the complexities of subject matter, potentially limiting the effectiveness of the LLM system. Providing structured pathways and guidance within the LLM framework could support students in navigating complex problems.

Proposed Solutions: Implementing structured guidance and support mechanisms within LLM systems could help students design effective prompts and engage with complex subject matter.

Project Team

Abdullah Mushtaq

Researcher

Muhammad Rafay Naeem

Researcher

Ibrahim Ghaznavi

Researcher

Muhammad Imran Taj

Researcher

Imran Hashmi

Researcher

Junaid Qadir

Researcher

Contact Information

For information about the paper, please contact the authors.

Authors: Abdullah Mushtaq, Muhammad Rafay Naeem, Ibrahim Ghaznavi, Muhammad Imran Taj, Imran Hashmi, Junaid Qadir

Source Publication: View Original PaperLink opens in a new window

Project Contact: Dr. Jianhua Yang

LLM Model Version: gpt-4o-mini-2024-07-18

Analysis Provider: Openai

Let us know you agree to cookies