EvoOpt-LLM: Evolving industrial optimization models with large language models

TL;DR

EvoOpt-LLM introduces a data-efficient, LLM-based framework for automating and enhancing industrial optimization modeling, including model construction, constraint injection, and variable pruning, with significant performance improvements on MILP problems.

Contribution

The paper presents EvoOpt-LLM, a novel LLM-based system that automates industrial optimization modeling tasks with high data efficiency and scalability, addressing limitations of existing methods.

Findings

Achieves 91% generation rate and 65.9% executability with only 3,000 samples.

Reliably injects constraints while preserving objectives in MILP models.

Enhances computational efficiency with a 0.56 F1 score on medium LP models.

Abstract

Optimization modeling via mixed-integer linear programming (MILP) is fundamental to industrial planning and scheduling, yet translating natural-language requirements into solver-executable models and maintaining them under evolving business rules remains highly expertise-intensive. While large language models (LLMs) offer promising avenues for automation, existing methods often suffer from low data efficiency, limited solver-level validity, and poor scalability to industrial-scale problems. To address these challenges, we present EvoOpt-LLM, a unified LLM-based framework supporting the full lifecycle of industrial optimization modeling, including automated model construction, dynamic business-constraint injection, and end-to-end variable pruning. Built on a 7B-parameter LLM and adapted via parameter-efficient LoRA fine-tuning, EvoOpt-LLM achieves a generation rate of 91% and an executability rate of 65.9% with only 3,000 training samples, with critical performance gains emerging under 1,500 samples. The constraint injection module reliably augments existing MILP models while preserving original objectives, and the variable pruning module enhances computational efficiency, achieving an F1 score of ~0.56 on medium-sized LP models with only 400 samples. EvoOpt-LLM demonstrates a practical, data-efficient approach to industrial optimization modeling, reducing reliance on expert intervention while improving adaptability and solver efficiency.