Abstract Two important enabling technologies for the Industrial Metaverse (IM) are the Digital Twin (DT) and Extended Reality (XR), an umbrella term for Virtual Reality (VR), Augmented Reality (AR) and Mixed Reality (MR). Projects and use cases that utilize these technologies simultaneously are those taking the next step toward Industrial Metaverse realization. With the concept still in its infancy, very little publicly available research has been conducted on such pioneering advances. The aim of this work is to summarize them and consider the factors which can prove to be challenges or opportunities for the Industrial Metaverse. Sustainability concerns proved to be both a challenge and an opportunity for pioneers and were analysed further.

Abstract As one of the key technologies of Industry 4.0, Digital Twin (DT) has been developed significantly since it was first proposed in 2002. Industry 5.0, as an extension and complement to Industry 4.0, focuses more on human-machine interaction. Human-in-the-loop (HiTL) can make DT more human-centric. However, the absence of human factors in DT has led to limited research in this area. Therefore, a systematic literature review of DT and HiTL is necessary. By reviewing and summarizing the development, definitions, characteristics, and applications of DT, as well as the theory, classification, and applications of HiTL systems, this study explores the potential for integrating DT and HiTL to realize DT for HiTL, while highlighting potential future research directions in this dynamic field. 

Abstract Large language models (LLMs) are increasingly used in design and manufacturing, yet directly employing general-purpose LLMs for conceptual design often leads to unmanufacturable concepts. This paper aims to adapt general-purpose LLMs for design-specific tasks. A new framework is presented to customize a general-purpose LLM into a design-specific model based on design-relevant data and Retrieval-Augmented Generation (RAG). Another complementary framework is presented to personalize the design-specific LLM by integrating design reasoning with prompting techniques. A design experiment, using patent documents as the design-relevant data, demonstrates that customization and personalization can improve LLM effectiveness in conceptual design, especially by enhancing concept feasibility.

Abstract As technology is evolving, newer and more advanced tools allow for the construction of more productive yet complex systems. Modern customers seek personalized products, creating a demand for highly flexible production. In an attempt for consumer products to become more competitive, it is necessary to reduce their cost. This means that fast and accurate decision-making is needed to avoid the production of defective products. This work presents a Digital Twin of a hybrid manufacturing cell which utilizes near real-time data and Artificial Intelligence in order to speed up the decision-making process of whether a product is defective before its production is complete. A real use case will be analysed to determine the feasibility of automating adaptive control in this manner.

Abstract Industry demands human‑centric, sustainable and resilient manufacturing, and next‑generation AI is reshaping how people work. This talk introduces a Human Digital Twin (HDT) framework for Human‑in‑the‑Loop Collaborative Manufacturing Systems (HiLoCS), addressing the lack of standardised guidelines for representing human needs and interactions. Building on ISO 23247, it classifies human needs into three hierarchical levels—safety, health and higher‑level—and systematically integrates them into the HDT. The framework decomposes HDT modelling into geometric, physical, behavioural, knowledge/rule and hybrid models, each capturing complementary aspects of human characteristics. A case study demonstrates how this approach can map human needs to system services and improve collaboration in human-robot collaborative assembly scenarios. The goal is to provide researchers with a scalable methodology for designing and deploying human‑centric collaborative systems that leverage both human intuition and digital tools. By linking these elements, the framework lays a foundation for continuous improvement of HiLoCS as our understanding of human factors evolves.