Benchmarking Open-endedness of VR Training Systems in Manufacturing

As conventional hands-on workshop training for manufacturing skills like welding and drilling is difficult to scale, virtual reality (VR) environments are being explored as a scalable industry training alternative due to their potential to provide immersive and interactive environments to practice and perfect psychomotor skills. Recent research on designing VR training for manufacturing has further demonstrated that an open-ended approach that allows multiple pathways to achieve tasks can improve training effectiveness. To fully harness the potential of open-ended VR training, it is essential to first study the current state of openness in existing systems. This paper addresses this need by conducting a benchmarking study of 19 industry applications evaluated on seven dimensions of open-endedness, including goal selection and multiple learning paths. Our findings reveal that current systems are moderately open-ended for goal selection and system adaptability, but have significant limitations in open-endedness for product creation, feedback mechanisms, and learning paths. Our analysis highlights design opportunities to further improve open-ended learning in VR training systems for machining, welding, and additive manufacturing.