Högskolan i Skövde

Digital technologies are critical in advancing a human-centered approach to warehouses that account for productivity and staff well-being. These technologies generate data addressing the negative conditions affecting the well-being of staff during order picking (OP), a labour intensive activity. This study analyzes artificial intelligence (AI)-enabled vision systems to enhance human-centricity and improve the generation and analysis of information about tasks executed by staff in OP. The study presents results from a pilot study in automotive manufacturing applying a design science research approach. The results show that AI-enabled vision systems enhance task identification, analysis, and efficiency in OP. The study suggests five actions including staff information, data acquisition, access restriction, data storage, and protection addressing the privacy concerns of these systems. The study discusses how these systems can integrate staff well-being by identifying human factors and outcomes. It offers three contributions: (1) an overview of activities for collecting task information through AI-enabled vision systems in human-centered OP; (2) evidence that existing architectures for human-centered manufacturing are essential for managing privacy implications; and (3) a discussion of the systems' impact on human factors and performance, and guidelines for developing and implementing these systems in future studies and operational environments.

Advanced digital technologies are developed and implemented to address challenges in complex manufacturing systems. Two well-known technologies in this domain are virtual commissioning and discrete-event modeling and simulation. These simulation tools are supported by well-substantiated theoretical frameworks and proven in industrial applications. However, the two approaches target different aspects of production systems and their combination is not extensively researched, thereby motivating this study. This paper reviews existing literature on the combination of these simulation approaches, aiming to identify current opportunities, challenges, and research gaps, while proposing innovative directions for future exploration informed by the findings. The results suggest future research should explore combining VC and DES to enable test-driven development, reduce modeling effort through joint model design, and extend simulation model usability across the production system life cycle.

Transitioning from linear to circular production is vital for manufacturing companies aiming to reduce environmental impact. Simultaneously, they face rapid technological change, fluctuating demands, and frequent product introductions. Addressing these challenges requires rethinking how production systems are designed and operated, not only to improve resource efficiency but also to extend system lifespan through relevance and resilience. Reconfigurability offers a promising approach, enabling systems to adapt continuously to evolving needs. However, literature lacks guidance on managing production systems that remain resource-efficient and long-lived, i.e., circular production systems. This study explores how circularity and reconfigurability can be operationalized across the lifecycle of production systems based on examples from six companies. It presents a spectrum of approaches, including technical solutions, support structures, and collaboration models. Technical solutions show how systems can be reused and reconfigured to maximize efficiency. Collaboration models emphasize revising design practices to foster partnerships between suppliers and manufacturers. Support structures offer insights into leveraging data to relocate and reuse components, avoiding premature decommissioning. Efficient reselling practices at end-of-life are also discussed. Overall, the paper bridges theory and practice, offering actionable strategies for industry and research.

This paper presents the design and development of X-MAIN (eXtended Reality for Maintenance and Inspection Training), an Extended Reality (XR) simulation framework for preventive maintenance, inspection, and safety verification of industrial robotics. The platform enables trainees to practice tool-based procedures and diagnostic workflows in a safe and repeatable virtual workshop, reducing the dependency on physical equipment and instructor availability. The framework is structured into two interconnected spaces: a Simulation Zone for operational and inspection activities (e.g., safety checks, calibration, sensor diagnostics, vibration analysis) and a Maintenance Zone for hands-on procedures (e.g., lubrication, component replacement, cleaning). A key contribution is a unified pair-based interaction model that formalizes user actions as tool–target pairs, allowing consistent evaluation and device-agnostic interaction across Virtual Reality (VR) and desktop deployments. The system integrates multi-modal feedback and logs performance metrics (time, errors, tool usage, safety violations) to support student/user assessment studies.

This paper presents the design and development of X-RAPT (eXtended Reality Adaptive Programming Training), a simulation framework aimed at enhancing robotics education through immersive extended reality (XR) technologies. X-RAPT is a multi-user platform that enables students to learn and practice industrial robotic programming in safe, virtual environments. The system combines real-time VR interaction, hand-tracking, voice recognition, and a web-based task editor to provide an adaptive and scalable learning experience. Using realistic 3D environments and virtual replicas of industrial robots such as the UR3 and ABB IRB120, X-RAPT allows learners to engage in training scenarios that simulate tasks like object assembly, labeling, and packaging. The platform supports collaborative sessions with role-based functionality and cross-device synchronization, enabling intuitive interaction and immediate feedback. Preliminary results indicate that X-RAPT is an effective tool for lowering the entry barrier to robotics programming and fostering hands-on learning in virtual training environments.

The increasing complexity of modern automotive lighting systems requires advanced validation strategies that ensure both functional performance and regulatory compliance. This study presents a structured integration of Software-in-the-Loop (SiL) and Hardware-in-the-Loop (HiL) testing within a digital twin (DT) framework for validating headlamp systems. A gated validation process (G10–G120) is proposed, aligning each development phase with corresponding simulation stages from early requirements and concept validation to real-world scenario testing and continuous integration. A key principle of this approach is the adoption of a framework built upon the V-Cycle, adapted to integrate DT technology with SiL and HiL workflows. This architectural configuration ensures a continuous data flow between the physical system, the digital twin, and embedded software components, enabling real-time feedback, iterative model refinement, and traceable system verification throughout the development lifecycle. The paper also explores strategies for effective DT integration, such as digital twin-as-a-service, which combines virtual testing with physical validation to support earlier fault detection, streamlined simulation workflows, and reduced dependency on physical prototypes during lighting system development. Unlike the existing literature, which often treats SiL, HiL, and DTs in isolation, this work proposes a unified, domain-specific validation framework. The methodology addresses a critical gap by aligning simulation-based testing with development milestones and regulatory standards, offering a foundation for industrial adoption.

With the emergence of Industry 5.0 and an increasing focus on human-centric approaches in manufacturing, the analysis of workers in production systems has gathered significant interest among researchers and practitioners. Previous studies have explored the impact of various aspects, such as skills, fatigue, and circadian rhythms, on human performance. However, the cumulative effect of these aspects as disturbances on work performance has yet to be fully elucidated. This study introduces an approach using the Functional Resonance Analysis Method (FRAM) to investigate the impact of multiple disturbances on workers’ performance. Furthermore, this approach explored how the resilience-related skill aspects of workers affect their performance under multiple disturbances. A case study on engine test and repair processes was conducted, employing qualitative data collection and semi-quantitative simulation studies examining the impact of combined disturbances across 4,094 scenarios. The results show that a larger number of compounded variabilities expressed in Common Performance Conditions (CPCs) made it significantly challenging to recover work performance, and CPCs with particularly critical effects were identified. In addition, the FRAM model of skilled workers was shown to sustain higher performance across more scenarios. The approach of this study has demonstrated its ability to provide insights for effectively and safely managing production systems while considering complex disturbances.

Dynamic order picking has usually demonstrated significant impacts on production efficiency in warehouse management. In the context of an automotive-part warehouse, this paper addresses a dynamic multi-tour order-picking problem based on a novel attention-aware deep reinforcement learning-based (ADRL) method. The multi-tour represents that one order-picking task must be split into multiple tours due to the cart capacity and the operator’s workload constraints. First, the multi-tour order-picking problem is formulated as a mathematical model, and then reformulated as a Markov decision process. Second, a novel DRL-based method is proposed to solve it effectively. Compared to the existing DRL-based methods, this approach employs multi-head attention to perceive warehouse situations. Additionally, three improvements are proposed to further strengthen the solution quality and generalization, including (1) the extra location representation to align the batch length during training, (2) the dynamic decoding to integrate real-time information of the warehouse environment during inference, and (3) the proximal policy optimization with entropy bonus to facilitate action exploration during training. Finally, comparison experiments based on thousands of order-picking instances from the Swedish warehouse validated that the proposed ADRL could outperform the other twelve DRL-based methods at most by 40.6%, considering the optimization objective. Furthermore, the performance gap between ADRL and seven evolutionary algorithms is controlled within 3%, while ADRL can be hundreds or thousands of times faster than these EAs regarding the solving speed.

Extended reality (XR) technologies are gaining traction in technical education due to their potential for creating immersive and interactive training environments. This study presents the development and empirical evaluation of X-RAPT, a collaborative VR-based platform designed to train students in industrial robotics programming. The system enables multi-user interaction, cross-platform compatibility (VR and PC), and real-time data logging through a modular simulation framework. A pilot evaluation was conducted in a vocational training institute with 15 students performing progressively complex tasks in alternating roles using both VR and PC interfaces. Performance metrics were captured automatically from system logs, while post-task questionnaires assessed usability, comfort, and interaction quality. The findings indicate high user engagement and a distinct learning curve, evidenced by progressively shorter task completion times across levels of increasing complexity. Role-based differences were observed, with main users showing greater interaction frequency but both roles contributing meaningfully. Although hardware demands and institutional constraints limited the scale of the pilot, the findings support the platform’s potential for enhancing robotics education.

This chapter examines integrating the Functional Resonance Analysis Method (FRAM) with Discrete-Event Simulation (DES) to analyze and enhance socio-technical production systems in the context of Industry 5.0 and Society 5.0. It highlights the complexity of interactions among technologies, humans, and societal elements, emphasizing FRAM’s potential for quantitative analysis and its combination with DES for system improvement, risk prevention, and resilience. Practical case studies demonstrate these methods’ applications, illustrating their contributions to system performance and risk assessment, paving the way for more advanced industries and societies.