How to make nuclear fusion “XR-ready”: A role-based CAD-to-XR framework

While interest in Extended Reality (XR) applications within the nuclear fusion community is rapidly increasing, the lack of structured and reproducible CAD-to-XR frameworks still limits their systematic adoption. Existing approaches are often fragmented, focusing on isolated optimization techniques without providing explicit decision-making support or coordination among the different professional roles involved in the pipeline. This work proposes a two-phase CAD-to-XR framework grounded in a theatrical metaphor, in which the XR experience is conceived as a stage play and 3D models are treated as actors with different narrative and functional relevance. In Phase 1, the framework supports experience design and optimization planning through a structured role-assignment process, while in Phase 2 it operationalizes these decisions through role-based 3D model optimization and iterative assessment in XR, combining objective performance metrics and expert qualitative assessment. The framework is illustrated through its application to the Divertor Tokamak Test (DTT) case study, showing its potential to support collaboration among XR designers, 3D modelers, XR developers, and domain experts, while facilitating CAD-to-XR transformation on standalone head-mounted devices. The proposed approach offers a practical decision-support framework, aiming at reducing CAD-to-XR pipeline inefficiencies and supporting the adoption of immersive XR technologies in nuclear fusion workflows.