INTEGRATING COMPUTER-AIDED ENGINEERING AND COMPUTER VISION FOR HUMAN-CENTRED WORKSPACE DESIGN IN INDUSTRY 5.0: A SYSTEMATIC LITERATURE REVIEW

Industry 4.0 has morphed into Industry 5.0 and shifted the focus from a technology-only approach to manufacturing to one that considers human-centricity, sustainability and resilience. This is transforming the design of manufacturing work spaces in the ways engineers can capture the physical workspace but can also capture the posture, movement and capabilities of the people who work within the space. This picture can be broken down into several components: Computer-Aided Engineering (CAE), Finite Element Analysis (FEA), Digital Human Modelling (DHM), computer vision, ergonomic assessment, human–robot collaboration (HRC) and Human Digital Twins (HDTs) all play a part in this. The aim of this review is to analyze the current state of the integration and contribution of these technologies to human-centred workspace design in the context of Industry 5.0, according to the PRISMA 2020 guidelines, 39 peer-reviewed studies published since 2016 were synthesized. As it becomes clear: CAE and DHM can be used to test and fine-tune a workspace before it is constructed, and computer vision enables the actual way people move and hold themselves in the shop to be captured. When combined, the two allow for automated ergonomic risk assessment, more optimised work stations, HRC customized to the individual and continuous real-time monitoring, while HDTs take this one step further by connecting a physical worker to a virtual worker, that informs manufacturing decisions. Real obstacles remain. Some of the remaining open questions are how to achieve interoperability, how to validate computer vision, how to maintain privacy, how to process data in real-time, and how to meaningfully embed the human body and mind in engineering models, which are still confronted by issues of accuracy, occlusion, and placement of the camera. The evidence from the emerging industrial economies is particularly light, where the adoption of digitalisation is patchy and resources are limited. The review suggests an integrated eight-layer structure in which the CAE, computer vision, HDTs and statistically based optimisation will combine in closed loop to continually modify workspaces to match the people using them, as Industry 5.0 demands human-centricity.