.png){kind=icon}
返回
Shopfloor layout generation method based on large language models
.png){kind=icon}
PRE
AI
DOI：10.1080/0951192X.2025.2482258.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Traditional shopfloor layout design faces limitations such as high complexity in model construction and excessive reliance on individual designers' experience, which cannot meet the diverse needs of modern industrial production. To obtain solutions with adaptive and customizable potential, this paper introduces a current cutting-edge technology - Large Language Models (LLMs). The intention is to leverage their spatial reasoning and generative capabilities to open up new avenues for solving such problems. This paper proposes a shopfloor layout generation method based on Prompt Engineering. By structurally defining and semantically transforming the layout space and task requirements, integrating the multi-dimensional information of the shopfloor as the knowledge support of the generation process, and constructing an iterative cycle of 'think-generate-evaluate', a stable generation of effective shopfloor layouts is achieved. This paper constructed a specialized Q&A dataset to evaluate the boundaries and potential of large language models in the field of shopfloor layout. Additionally, an experimental validation was conducted on three representative layout patterns in typical discrete manufacturing shopfloor scenarios, strongly demonstrating the innovative value and practical significance of the method.
Keyword:
Large language models
shopfloor layout
prompt engineering
期刊
I
IF:
4
论文数:
2.3K
被引数:
3.4K
机构
引用论文
Embodied intelligence in manufacturing: leveraging large language models for autonomous industrial robotics制造业中的具体化智能: 利用大型语言模型实现自主工业机器人
PREAI
分享收藏Enhancing Zero-Shot Crypto Sentiment With Fine-Tuned Language Model and Prompt Engineering.png)
IEEE ACCESS
OAAI
IEEE ACCESSIF3.6
OAAI分享收藏MED-Prompt: A novel prompt engineering framework for medicine prediction on free-text clinical notes
OAAI
OAAI分享收藏
查看更多内容