This research introduces Planning of Heuristics (PoH), a novel method leveraging large language models to automatically discover effective heuristics for complex optimization problems without human expertise.

• Integrates Monte Carlo Tree Search with LLMs to strategically explore the heuristic solution space
• Eliminates the need for extensive domain knowledge in designing optimization heuristics
• Demonstrates effective performance on classical problems like Traveling Salesman Problem
• Creates a framework where AI can autonomously develop and refine optimization strategies

This innovation matters for Engineering by providing a systematic approach to automate the discovery of optimization techniques across various domains, potentially accelerating solution development for complex industrial problems.

Planning of Heuristics: Strategic Planning on Large Language Models with Monte Carlo Tree Search for Automating Heuristic Optimization