Enabling interactive AI in industry 5.0 with RAG-enhanced GenAI Chatbots

Zhou, Tianyu, Wan, Yuwei, Liu, Ying ORCID: https://orcid.org/0000-0001-9319-5940 and Kumar, Maneesh ORCID: https://orcid.org/0000-0002-2469-1382 2025. Enabling interactive AI in industry 5.0 with RAG-enhanced GenAI Chatbots. Presented at: 31st International Conference on Engineering, Technology, and Innovation- ICE IEEE/ITMC, Valencia, Spain, 16-19 June 2025.

Preview

PDF - Presentation Download (887kB) | Preview

Abstract

Industry 5.0 advances sustainable development through human-machine collaboration and personalised manufacturing. The increase in intelligent industrial equipment creates data scalability challenges for human workers who face difficulties in making decisions based on relevant data sources. Advanced interactive AI systems, capable of integrating diverse data sources and delivering real-time, context-aware insights, present promising solutions to the challenges of the industrial environment. This research introduces a retrieval-augmented generation (RAG)-enhanced Generative artificial intelligence (GenAI) chatbot to address these challenges. The system integrates a variety of information sources, including government reports, news websites, academic studies, and industry reports. This industry 5.0 chatbot aims to offer users extensive knowledge of the industrial sector through a Question-and-Answer interface. It provides relevant and accurate information through intuitive, context-aware interactions to reduce cognitive load for users, which improves decision-making efficiency and user experience. Through experimental evaluation, the RAG-enhanced GenAI chatbot significantly improves accuracy, relevance and user satisfaction, outperforming models like ChatGPT-4o. This system presents an innovative practical solution to tackle Industry 5.0 core issues particularly in enhancing human-machine collaboration and decision-making efficiency. This research contributes to the theoretical and practical development of RAG-enhanced AI systems, laying a foundation for future investigations of industrial AI interaction.

Item Type:	Conference or Workshop Item (Paper)
Status:	Unpublished
Schools:	Schools > Business (Including Economics) Schools > Engineering
Last Modified:	26 Jun 2025 15:15
URI:	https://orca.cardiff.ac.uk/id/eprint/178617

Actions (repository staff only)

Edit Item