A review on cognitive workload for industry 5.0

In the evolving landscape of Industry 5.0 (I5.0), where digital technologies are increasingly integrated into industrial processes, understanding cognitive workload (CWL) during maintenance tasks has become critical. CWL significantly influences an operator's performance, safety, and overall well-being, especially in complex and demanding environments. The introduction of cognitive and assistive technologies, such as augmented reality (AR), virtual reality (VR), and artificial intelligence, holds the potential for reducing cognitive strain. However, existing research largely focuses on post-hoc CWL assessment rather than on integrating CWL considerations into the design phase of maintenance systems, according to an I5.0 perspective. Additionally, methodologies for accurately measuring and modelling CWL in real-time remain underdeveloped. In this context, assessing the operator's CWL can be a key factor in evaluating design and management alternatives for industrial systems, aiming to ensure the operator's well-being. Reducing CWL should, therefore, be a criterion for evaluating maintenance systems, including task execution methods and how support information is presented. This study addresses these gaps by investigating, through a systematic literature review, the existing methods to evaluate operators’ CWL and explores how they can be integrated into managing maintenance operations in the I5.0 context, with a specific focus on scenarios where digital technologies provide support. The identified CWL assessment approaches were categorised into three primary areas: CWL as a factor influencing the operator's performance, CWL as a measure for assessing the effectiveness of solutions, and CWL as a design driver. The findings reveal that AR and VR applications are widely adopted for supporting maintenance activities, but there are no clear results on their potential to reduce the operator's CWL. Moreover, results indicate that practical methodologies for real-time CWL monitoring and predictive modelling are lacking. We highlight the need for robust models to minimize CWL based on task and environmental factors, aligning with I5.0′s emphasis on human-centred design. The study contributes to the body of knowledge by identifying key research gaps and proposing a structured framework for CWL assessment in industrial systems development. It emphasizes the development of integrative methodologies for CWL assessment that are based on both subjective and physiological measurements. It moreover offers practical insights for designing maintenance systems that prioritize operator cognitive well-being alongside performance efficiency.