Intent-Driven Smart Manufacturing Integrating Knowledge Graphs and Large Language Models

TL;DR

This paper introduces a framework combining instruction-tuned large language models with knowledge graphs to translate human intents into machine actions in manufacturing, improving accuracy and operational alignment.

Contribution

It presents a novel integration of LLMs and knowledge graphs for intent-driven manufacturing, with domain-specific fine-tuning and semantic mapping to enhance human-machine interaction.

Findings

Achieved 89.33% exact match accuracy in intent translation

Demonstrated significant performance improvements over baseline models

Enabled semantic alignment with manufacturing standards like ISA-95

Abstract

The increasing complexity of smart manufacturing environments demands interfaces that can translate high-level human intents into machine-executable actions. This paper presents a unified framework that integrates instruction-tuned Large Language Models (LLMs) with ontology-aligned Knowledge Graphs (KGs) to enable intent-driven interaction in Manufacturing-as-a-Service (MaaS) ecosystems. We fine-tune Mistral-7B-Instruct-V02 on a domain-specific dataset, enabling the translation of natural language intents into structured JSON requirement models. These models are semantically mapped to a Neo4j-based knowledge graph grounded in the ISA-95 standard, ensuring operational alignment with manufacturing processes, resources, and constraints. Our experimental results demonstrate significant performance gains over zero-shot and 3-shots baselines, achieving 89.33\% exact match accuracy and 97.27\% overall accuracy. This work lays the foundation for scalable, explainable, and adaptive human-machine