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

2026-02-01

AI Summary

This work addresses the challenge of translating natural language requirements into executable mixed-integer linear programming (MILP) models in industrial optimization, a process traditionally reliant on expert knowledge. Existing large language model (LLM) approaches suffer from low data efficiency, poor solver compatibility, and limited scalability. To overcome these limitations, we propose the first unified LLM framework that supports dynamic evolution and efficient maintenance of MILP models, integrating automated modeling, dynamic constraint injection, and end-to-end variable pruning. Using a 7B-parameter model with LoRA fine-tuning, our approach achieves a 91% generation rate and 65.9% executability rate with only 3,000 training samples. The variable pruning module attains an F1 score of 0.56 on medium-scale LP instances using just 400 samples, significantly improving solver efficiency while ensuring injected constraints preserve the original objective.

Technology Category

Machine Learning: Large Multimodal Models (LMMs)Natural Language Processing: (Large) Language ModelsPlanning, Routing, and Scheduling: Planning with Language Models

Application Category

User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendationEconomics, Online Markets and Human Computation: Cost models of using LLMs in production systemsGraph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

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.

Problem

• optimization modeling
• mixed-integer linear programming
• large language models
• industrial planning
• business rule evolution

Innovation

• LLM-based optimization
• MILP modeling
• constraint injection
• variable pruning
• LoRA fine-tuning