Automatic programming via large language models with population self-evolution for dynamic job shop scheduling problem

TL;DR

This paper introduces a novel population self-evolutionary framework that leverages large language models to automatically design heuristic dispatching rules for dynamic job shop scheduling, outperforming existing methods especially in unseen scenarios.

Contribution

The paper presents the SeEvo method, a new framework combining LLMs and evolutionary strategies for automatic HDR design, improving search efficiency and generalization in dynamic environments.

Findings

SeEvo outperforms GP, GEP, and deep reinforcement learning methods.

SeEvo surpasses over 10 common heuristic dispatching rules.

SeEvo demonstrates superior performance in unseen and dynamic scenarios.

Abstract

Heuristic dispatching rules (HDRs) are widely regarded as effective methods for solving dynamic job shop scheduling problems (DJSSP) in real-world production environments. However, their performance is highly scenario-dependent, often requiring expert customization. To address this, genetic programming (GP) and gene expression programming (GEP) have been extensively used for automatic algorithm design. Nevertheless, these approaches often face challenges due to high randomness in the search process and limited generalization ability, hindering the application of trained dispatching rules to new scenarios or dynamic environments. Recently, the integration of large language models (LLMs) with evolutionary algorithms has opened new avenues for prompt engineering and automatic algorithm design. To enhance the capabilities of LLMs in automatic HDRs design, this paper proposes a novel population self-evolutionary (SeEvo) method, a general search framework inspired by the self-reflective design strategies of human experts. The SeEvo method accelerates the search process and enhances exploration capabilities. Experimental results show that the proposed SeEvo method outperforms GP, GEP, end-to-end deep reinforcement learning methods, and more than 10 common HDRs from the literature, particularly in unseen and dynamic scenarios.