Automatic MILP Model Construction for Multi-Robot Task Allocation and Scheduling Based on Large Language Models

TL;DR

The paper tackles privacy-sensitive multi-robot task allocation and scheduling by introducing a knowledge-augmented MILP framework that leverages local LLMs and domain knowledge to convert natural language descriptions into executable MILP code, solved by Gurobi. It introduces a two-phase pipeline: (1) fuzzy NL descriptions are converted into precise MILP models via a knowledge-base-guided local LLM, and (2) those MILP models are translated into code using a template-based approach and a supervised fine-tuned local LLM, with validation on a private dataset. Key contributions include an 82% average accuracy in constraint extraction and a 90% average code-generation accuracy after SFT, supported by a large SFT dataset (~10k entries with 494 verified), and demonstrated in an aircraft skin manufacturing case under data privacy constraints. The results suggest a practical, low-cost path for automatic modeling and solution of complex MRTA problems in industry, with future work aimed at improving efficiency and extending to broader workshop scheduling scenarios.

Abstract

With the accelerated development of Industry 4.0, intelligent manufacturing systems increasingly require efficient task allocation and scheduling in multi-robot systems. However, existing methods rely on domain expertise and face challenges in adapting to dynamic production constraints. Additionally, enterprises have high privacy requirements for production scheduling data, which prevents the use of cloud-based large language models (LLMs) for solution development. To address these challenges, there is an urgent need for an automated modeling solution that meets data privacy requirements. This study proposes a knowledge-augmented mixed integer linear programming (MILP) automated formulation framework, integrating local LLMs with domain-specific knowledge bases to generate executable code from natural language descriptions automatically. The framework employs a knowledge-guided DeepSeek-R1-Distill-Qwen-32B model to extract complex spatiotemporal constraints (82% average accuracy) and leverages a supervised fine-tuned Qwen2.5-Coder-7B-Instruct model for efficient MILP code generation (90% average accuracy). Experimental results demonstrate that the framework successfully achieves automatic modeling in the aircraft skin manufacturing case while ensuring data privacy and computational efficiency. This research provides a low-barrier and highly reliable technical path for modeling in complex industrial scenarios.

Figures (4)

• Figure 1: LLM-Augmented Framework for Multi-Robot Task Allocation and Scheduling.
• Figure 2: The knowledge-augmented automated MILP formulation framework.
• Figure 3: SFT dataset construction pipeline.
• Figure 4: Result for cases automatically generated by knowledge-augmented automated MILP formulation framework.