REMoH: A Reflective Evolution of Multi-objective Heuristics approach via Large Language Models

TL;DR

REMoH introduces a novel framework combining NSGA-II with Large Language Models to enhance multi-objective heuristics through reflection, clustering, and search-space exploration, leading to improved convergence and diversity in complex scheduling problems.

Contribution

The paper presents REMoH, a new LLM-integrated heuristic evolution framework that improves multi-objective optimization by incorporating reflection mechanisms for heuristic generation.

Findings

REMoH achieves competitive results on FJSSP benchmarks.

The approach reduces modeling effort compared to traditional methods.

REMoH enhances adaptability and solution diversity in multi-objective problems.

Abstract

Multi-objective optimization is fundamental in complex decision-making tasks. Traditional algorithms, while effective, often demand extensive problem-specific modeling and struggle to adapt to nonlinear structures. Recent advances in Large Language Models (LLMs) offer enhanced explainability, adaptability, and reasoning. This work proposes Reflective Evolution of Multi-objective Heuristics (REMoH), a novel framework integrating NSGA-II with LLM-based heuristic generation. A key innovation is a reflection mechanism that uses clustering and search-space reflection to guide the creation of diverse, high-quality heuristics, improving convergence and maintaining solution diversity. The approach is evaluated on the Flexible Job Shop Scheduling Problem (FJSSP) in-depth benchmarking against state-of-the-art methods using three instance datasets: Dauzere, Barnes, and Brandimarte. Results demonstrate that REMoH achieves competitive results compared to state-of-the-art approaches with reduced modeling effort and enhanced adaptability. These findings underscore the potential of LLMs to augment traditional optimization, offering greater flexibility, interpretability, and robustness in multi-objective scenarios.