Publications
Journals
- A. Sbaragli, M. Nardello ‘From Sensing to Segmentation: Transformer-Based Worker Activity Recognition for Industrial Assembly’, IEEE Access, vol. 13, no. 19, pp. 207475--207487, 2025. doi: 10.1109/ACCESS.2025.3641299.
Abstract ▼ BibTeX ▼
Abstract
Human operators still provide strategic, value-added contributions to modern manufacturing systems, despite the widespread adoption of automation. This paper presents a cyber-physical system designed to perform fine-grained Worker Activity Recognition (WAR) and segment manual operations in human-centric assembly environments. The proposed system includes a modular and low-intrusion IoT acquisition layer that digitizes worker movements using Inertial Measurement Units (IMUs) and a radio frequency-based smart glove. While the IMUs track hand and back motion, the glove captures process interactions such as picking and depositing components. These multimodal data streams are processed by a cyber layer that performs time- and frequency-domain feature extraction and feeds the data into a Transformer-based neural network to classify five common assembly tasks. The results are post-processed by a Decision Support System to generate interpretable task segments and provide key performance indicators for industrial supervisors. The system is validated through a real-world deployment involving three workers assembling a vertical centrifugal pump, resulting in over 7 hours of annotated data. The Transformer model achieves an F1-score of 82% on this dataset. Thanks to its unobtrusive sensors, low deployment cost, and robustness to occlusions, the proposed solution offers a practical and scalable approach to enhancing visibility and traceability in manual industrial operations.
BibTeX
@article{sbaragli2025sensing, title={From Sensing to Segmentation: Transformer-Based Worker Activity Recognition for Industrial Assembly}, author={Sbaragli, Andrea and Nardello, Matteo}, journal={IEEE Access}, volume={13}, pages={207475--207487}, year={2025}, publisher={IEEE} } - A. Sbaragli, P. Y. Ghafoorpoor, S. Thiede, F. Pilati ‘A cyber-physical architecture to monitor human-centric reconfigurable manufacturing systems’, Journal of Intelligent Manufacturing, pp. 1–23, 2025. doi: 10.1007/s10845-024-02558-1.
Abstract
Reconfigurable manufacturing systems represent the most adequate production paradigm due to their ability to meet mass customized needs while ensuring cost-effective flexibilities and performances. However, digital solutions are required to manage these dynamic environments over working shifts and processes’ reconfiguration. In this scenario, this work proposes a layout and task-insensitive cyber-physical architecture to monitor human-centric reconfigurable manufacturing systems. Workers’ motion patterns and industrial resources’ positions are acquired through a radio-frequency-based real-time locating system. These data streams are fed into a machine-learning cyber layer to segment operators’ activities during production cycles into two steps. The first computational stream assigns workers’ motion patterns to industrial resources regardless of the system configuration. The following step distinguishes workers’ operations into value-added and non-value-added. These outputs are stored in a decision support system where customized callback functions develop key performing indicators to monitor the performance of such reconfigurable human-centric environments. The validity of the cyber-physical architecture is demonstrated in an industrial-related pilot environment, involving 40 workers and 8 production set-ups.
BibTeX
@article{sbaragli2025cyber, title={A cyber-physical architecture to monitor human-centric reconfigurable manufacturing systems}, author={Sbaragli, Andrea and Ghafoorpoor, Poorya Yazdi and Thiede, Sebastian and Pilati, Francesco}, journal={Journal of Intelligent Manufacturing}, pages={1--23}, year={2025}, publisher={Springer}, doi={10.1007/s10845-024-02558-1} } - F. Pilati, A. Sbaragli, T. Ruppert, J. Abonyi ‘Goal-oriented clustering algorithm to monitor the efficiency of logistic processes through real-time locating systems’, International Journal of Computer Integrated Manufacturing, vol. 37, no. 10-11, pp. 1359–1375, 2024.
Abstract
Modern internal logistic systems face several challenges, from supply chain disruption to mass customization of marketed products. In such a highly dynamic scenario, Internet of Things technologies provide a reliable path to digitizing low-standardized systems and quantitatively monitoring their functioning. In addition, acquired measurements are often combined with machine learning methods to achieve improved data analytics. For this purpose, this work presents a digital architecture to detect logistic activities during order management. While an ultrawide band-based real-time locating system acquires the positioning information of forklifts, a goal-oriented clustering algorithm called Industrial DB scan classifies process-driven operations during the shift. These insights represent valuable information for constantly evaluating the operational efficiency of logistic systems. The robustness and validity of the industrial DB scan are tested from different perspectives. On the one hand, a quantitative benchmark with traditional clustering methods is performed. The proposed algorithm results in the most effective approach to detect uptime forklift operations. On the other hand, a warehousing system proves the operational functioning of the algorithm. In this regard, a Tracking Management System interface is developed to achieve user-friendly process monitoring, where plant supervisors can analyze several internal logistic key performance indicators.
BibTeX
@article{pilati2024goal, title={Goal-oriented clustering algorithm to monitor the efficiency of logistic processes through real-time locating systems}, author={Pilati, Francesco and Sbaragli, Andrea and Ruppert, Tam{\'a}s and Abonyi, J{\'a}nos}, journal={International Journal of Computer Integrated Manufacturing}, volume={37}, number={10-11}, pages={1359--1375}, year={2024}, publisher={Taylor \& Francis} } - F. Tomelleri, A. Sbaragli, F. Picariello, F. Pilati ‘Digital ergonomic assessment to enhance the physical resilience of human-centric manufacturing systems in Industry 5.0’, Journal of Manufacturing Systems, vol. 77, pp. 246–265, 2024.
Abstract
The emergence of Industry 5.0 promotes the creation of human-centric values. To fulfill this objective, Internet of Things (IoT) technologies are increasingly being exploited to digitize the human factor and monitor the ergonomics of manual manufacturing systems. These digital assessments, combined with computational algorithms, contribute to the establishment of socially inclusive workplaces while offering detailed insights to safeguard the health of the aging workforce. In this scenario, this study proposes a digital architecture for evaluating the European Assembly Worksheet (EAWS) in human-centric manufacturing systems. Three distinct enabling technologies are leveraged to acquire heterogeneous data streams. A radio-frequency-based smart glove detects the operator’s interactions with the surrounding environment, while a network of marker-less cameras and a four-channel surface Electromyography (sEMG) system capture body joint movements and muscular contractions of the upper limbs, respectively. The acquired data are processed by computational algorithms to define an EAWS-driven set of Key Risk Indicators (KRIs), embedded in an ergonomic decision support system. These risk metrics highlight operator-driven process weaknesses in musculoskeletal, muscular, and material handling dimensions. Finally, the validity of the proposed digital architecture is demonstrated in an industrial-related pilot environment, where an operator assembles a piece of home furniture.
BibTeX
@article{tomelleri2024digital, title={Digital ergonomic assessment to enhance the physical resilience of human-centric manufacturing systems in Industry 5.0}, author={Tomelleri, Federica and Sbaragli, Andrea and Picariello, Francesco and Pilati, Francesco}, journal={Journal of Manufacturing Systems}, volume={77}, pages={246--265}, year={2024}, publisher={Elsevier} } - F. Pilati, A. Sbaragli ‘Learning human-process interaction in manual manufacturing job shops through indoor positioning systems’, Computers in Industry, vol. 151, pp. 103984, 2023.
Abstract
Nowadays, manufacturing systems are increasingly embracing the Industry 4.0 paradigm. Therefore, manual and low-standardized manufacturing environments are often digitized through Industrial Internet of Things technologies to quantitatively assess and investigate the role of the human factor from multiple points of view. This approach is commonly known as Operator 4.0. In such a scenario, this manuscript proposes an original digital architecture to monitor the efficiency and the social sustainability of labor-intensive manufacturing job shops. While the anonymous spatio-temporal trajectories of tagged workers are acquired through an ultrawide band radio network, machine learning algorithms autonomously detect the human-process interactions with strategic industrial entities upon developing industrial key performing indicators. The proposed architecture is tested and validated in a real manual manufacturing system. In detail, the performing accuracies of the machine learning-based software provide industrial plant supervisors with several production metrics to identify the hidden weaknesses and bottlenecks of the monitored manufacturing system. Such digital assessment may trigger a re-organization of the considered process to, for instance, enhance the allocation of the material in storage areas while fairly re-balancing the distances traveled by workers for picking activities.
BibTeX
@article{pilati2023learning, title={Learning human-process interaction in manual manufacturing job shops through indoor positioning systems}, author={Pilati, Francesco and Sbaragli, Andrea}, journal={Computers in Industry}, volume={151}, pages={103984}, year={2023}, publisher={Elsevier} }
Conferences
- F. Tomelleri, A. Sbaragli, F. Piacariello, F. Pilati ‘Safe assembly in industry 5.0: Digital architecture for the ergonomic assembly worksheet’, Procedia CIRP, vol. 127, pp. 68–73, 2024. doi: .
Abstract
The evolving landscape of modern manufacturing is affected by a confluence of social challenges arising from demographic shifts and erratic market demands. The emergence of Industry 5.0 revolutionized practices, with a keen focus on harnessing the potential of the Internet of Things to digitize the workforce embracing a human-centric approach. In this research, an original digital architecture integrates sensors into manufacturing environments to evaluate the well-being of assembly operators. The core objective is to capture human-process interactions, empowering production managers to optimize assembly environments from the ergonomic perspective. Three key enabling technologies are employed: radio frequency identification smart gloves, motion capture cameras, and superficial electromyography sensors. This physical layer detects and analyzes assembly tasks, operator movements, and muscular activities. Computational algorithms mine these data streams to assess the Ergonomic Assembly Worksheet, automatically. In detail, the investigated sections of this ergonomic index are basic postures, action forces, and manual material handling. Furthermore, a supplementary set of Key Risk Indicators supports production managers in evaluating the physical resilience of the assembly systems. These operator-specific metrics provide strategic information to trigger workstation redesign, task rebalancing, and other corrective actions for enhancing the safety of whichever human-centric assembly process. Finally, an experimental campaign in a controlled industrial environment tests the architecture’s effectiveness and potential to reduce the risks and enhance workforce health.
BibTeX
@article{tomelleri2024safe, title={Safe assembly in industry 5.0: Digital architecture for the ergonomic assembly worksheet}, author={Tomelleri, Federica and Sbaragli, Andrea and Piacariello, Francesco and Pilati, Francesco}, journal={Procedia CIRP}, volume={127}, pages={68--73}, year={2024}, publisher={Elsevier} } - A. Sbaragli, F. Tomelleri, F. Picariello, E. Picariello, F. Pilati ‘Safe Operator 5.0 digital architecture: towards resilient human-centric manufacturing systems’, IFAC-PapersOnLine, vol. 58, no. 19, pp. 265–270, 2024. doi: .
Abstract
The current shift in European demographic pyramids and the increase in the retirement age are posing serious challenges to the manufacturing sector. To address these social challenges, the recent Industry 5.0 paradigm is promoting human-centric value creation to achieve resilient and socially sustainable industrial environments. In this evolving scenario, this work conceptualizes the Safe Operator 5.0 digital architecture to safeguard the system and physical resilience of human-centric manufacturing environments leveraging three main layers. While workers are digitized through positioning, proximity, and physiological sensors, a machine learning-based cyber layer processes these data streams as well as managerial data to evaluate the efficiency and physical well-being of the workforce. Finally, a central management system provides feedback and early warning to final users.
BibTeX
@article{sbaragli2024safe, title={Safe Operator 5.0 digital architecture: towards resilient human-centric manufacturing systems}, author={Sbaragli, Andrea and Tomelleri, Federica and Picariello, Francesco and Picariello, Enrico and Pilati, Francesco}, journal={IFAC-PapersOnLine}, volume={58}, number={19}, pages={265--270}, year={2024}, publisher={Elsevier} } - P. Y. Ghafoorpoor, A. Sbaragli, L. Peters, F. Pilati, S. Thiede ‘Cyber physical system for reconfigurable learning factories: Combining 3d simulations, reconfigurable layouts and real-time locating systems’, Conference on learning factories, pp. 28–35, 2024.
Abstract
Learning Factories are educational environments that replicate manufacturing environment operations, providing next-generation engineers with hands-on learning opportunities. University-based Learning Factories may encounter barriers in supporting a wide range of learning courses with different intended learning objectives and outcomes that impede their successful implementation and can prevent valuable learning experiences. Overcoming these barriers requires dedicated efforts to enhance the infrastructural capabilities of the Learning Factory to become multi-use, invoking better flexibility and reconfigurability of the layout besides utilization of value-enhancing technologies. For this purpose, this work investigates the learning opportunities to enable data-driven layout reconfigurations by fusing 3D simulations with Real Time Locating Systems (RTLS). The seamless integration of these digital solutions provides accurate tracking and modeling of whichever manufacturing entity. Mirroring physical systems configurations into digital environments triggers learners to better analyze process functioning and performances. This digital solution is even more effective into modern mass-customized markets, where batch-centered production processes require easy-to-deploy and iterative layout optimizations. The findings of this empirical study support the transformative potential of achieving reconfigurable layouts alongside RTLS and 3D simulation technologies within a Learning Factory. This integrated approach is not limited to enhancing capabilities and operational efficiency but also provides a wider and multi-dimensional array of educational experiences, bridging the gap between academia and an ever-evolving manufacturing landscape.
BibTeX
@inproceedings{ghafoorpoor2024cyber, title={Cyber physical system for reconfigurable learning factories: Combining 3d simulations, reconfigurable layouts and real-time locating systems}, author={Ghafoorpoor Yazdi, Poorya and Sbaragli, Andrea and Peters, Leon and Pilati, Francesco and Thiede, Sebastian}, booktitle={Conference on learning factories}, pages={28--35}, year={2024}, organization={Springer} } - L. De Vito, E. Picariello, F. Picariello, S. Rapuano, I. Tudosa, A. Sbaragli, F. Pilati ‘IoT-based system for monitoring the well-being of industrial operators through wearable devices’, 2024 IEEE International Symposium on Medical Measurements and Applications (MeMeA), pp. 1–6, 2024. doi: .
Abstract
This paper proposes an Internet of Things-based system for monitoring the well-being of operators working in assembly lines. The system consists of two wearable devices, the former for the acquisition of physiological signals and the latter for the recognition of the operator postures. Furthermore, an Indoor Positioning System (IPS) is used for the real-time tracking of operators, tools, and components involved in the process. A preliminary metrological characterization of the system is performed on the heart rate (HR) measurements provided by the wearable electrocardiograph against a patient simulator, and the orientation measurements provided by the Inertial Measurement Unit (IMU). Furthermore, the latency affecting the IPS is measured. To validate the correct functioning of the entire system, a preliminary experimental campaign is conducted where one operator assembles and disassembles an industrial pump.
BibTeX
@inproceedings{de2024iot, title={IoT-based system for monitoring the well-being of industrial operators through wearable devices}, author={De Vito, Luca and Picariello, Enrico and Picariello, Francesco and Rapuano, Sergio and Tudosa, Ioan and Sbaragli, Andrea and Pilati, Francesco}, booktitle={2024 IEEE International Symposium on Medical Measurements and Applications (MeMeA)}, pages={1--6}, year={2024}, organization={IEEE} } - F. Pilati, A. Sbaragli, F. Tomelleri, E. Picariello, F. Picariello, I. Tudosa, M. Nardello ‘Operator 5.0: Enhancing the physical resilience of workers in assembly lines’, 2023 IEEE International Workshop on Metrology for Industry 4.0 & IoT (MetroInd4.0&IoT), pp. 177–182, 2023. doi: .
Abstract
The human factor represents the most fragile and valuable resource in modern and low-standardized manufacturing environments. Indeed, the Operator 5.0 concept aims at achieving socially-inclusive workplaces by monitoring the well-being of workers during production cycles. To accomplish this challenging aim, this manuscript proposes a digital industrial Internet-of-Things architecture to monitor the physical resilience of Operator 5.0 in assembly lines. While a markerless motion capture camera is adopted to evaluate the ergonomic exposure, a superficial electromyography wearable acquires muscular contractions of upper limbs to perform a machine learning-based recognition of fatigue status. In this preliminary investigation, the main focus of the analysis is to digitize the European Assembly Worksheet to evaluate the worker’s postures during the assembly of home furniture. Exploiting such ergonomic measurements, a Monte Carlo-based sensitivity analysis is leveraged to evaluate the noise in bending scenarios. Finally, a reference system is leveraged to assess the measurement error of the motion capture camera.
BibTeX
@inproceedings{pilati2023operator, title={Operator 5.0: Enhancing the physical resilience of workers in assembly lines}, author={Pilati, Francesco and Sbaragli, Andrea and Tomelleri, Federica and Picariello, Enrico and Picariello, Francesco and Tudosa, Ioan and Nardello, Matteo}, booktitle={2023 IEEE International Workshop on Metrology for Industry 4.0 \& IoT (MetroInd4. 0\&IoT)}, pages={177--182}, year={2023}, organization={IEEE} } - F. Pilati, A. Sbaragli, G. P. R. Papini, P. Capuccini ‘An artificial neural network architecture to classify workers’ operations in manual production processes’, International Conference on Flexible Automation and Intelligent Manufacturing, pp. 805–812, 2023. doi: .
Abstract
The recent Industry 4.0 paradigm is disruptively changing the manufacturing landscape. Where fully automated settings are not feasible or economically viable, Industrial Internet of Things sensors are gaining traction due to their flexibility and affordable costs. In such a scenario emerges crescent attention to digitizing the human factor. Based on this, this manuscript proposes an integrated digital architecture in which a radio-frequency-based indoor positioning system is adopted to anonymously tag human operators. The highly unbalanced spatio-temporal dataset is fed into a recurrent neural network architecture that classifies without overfitting the manual picking/deposit activities in products’ stocking areas of a real and low standardized manufacturing job shop with a performance of 0.93.
BibTeX
@inproceedings{pilati2023artificial, title={An artificial neural network architecture to classify workers’ operations in manual production processes}, author={Pilati, Francesco and Sbaragli, Andrea and Papini, Gastone Pietro Rosati and Capuccini, Paolo}, booktitle={International conference on flexible automation and intelligent manufacturing}, pages={805--812}, year={2023}, organization={Springer} } - L. De Vito, E. Picariello, F. Picariello, I. Tudosa, A. Sbaragli, G. P. R. Papini, F. Pilati ‘Measurement system for operator 5.0: a learning fatigue recognition based on sEMG signals’, 2023 IEEE International Symposium on Medical Measurements and Applications (MeMeA), pp. 1–6, 2023. doi: .
Abstract
In this paper, a fatigue recognition system, based on a Machine Learning (ML) algorithm is presented. A wearable device is used to acquire the sEMG signals on a subject performing complex tasks, using tools, and components. Different features are utilized in order to train the ML classifier, namely: amplitude features, frequency features, and used tools and components. In order to verify the effectiveness of the proposed system, various features have been chosen to train the classifier, i.e., an ensemble bagging decision tree, and a preliminary experimental assessment is presented, where the F1-score is calculated. The results show that through the use of all the proposed features and with an optimization phase of the classifier, it is possible to reach an F1-score of 77.7 %.
BibTeX
@inproceedings{de2023measurement, title={Measurement system for operator 5.0: a learning fatigue recognition based on semg signals}, author={De Vito, Luca and Picariello, Enrico and Picariello, Francesco and Tudosa, Ioan and Sbaragli, Andrea and Papini, Gastone Pietro Rosati and Pilati, Francesco}, booktitle={2023 IEEE International Symposium on Medical Measurements and Applications (MeMeA)}, pages={1--6}, year={2023}, organization={IEEE} } - M. Wolf, M. Rantschl, E. Auberger, H. Preising, A. Sbaragli, F. Pilati, C. Ramsauer ‘Real time locating systems for human centered production planning and monitoring’, IFAC-PapersOnLine, vol. 55, no. 2, pp. 366–371, 2022. doi: .
Abstract
Production companies often operate in a dynamic and volatile environment. This leads to an increasing demand for continuous changes in their production systems and processes. Furthermore, decreasing product life cycle times and rising market demand for product variety and individualized products is bringing about the necessity for the monitoring and coordination of processes in the operations phase. As a result of these developments production management is facing new challenges in decision making for the optimal settings of the production system design and the related coordination of production. All of this demands enormous efforts to maintain a consistent and reliable database for the ongoing configuration and coordination of the production system. It is thus a remarkable challenge for industrial companies. Real time locating systems (RTLS) with their ability to continuously monitor the current position and parameters (speed, direction, etc.) of process resources (operators, equipment, products, etc.) offer several potential benefits for the manufacturing industry. The potential areas of application can be identified in different layers of production management. They range from data acquisition on shop floor level through reconfiguration of production systems to providing real time feedback for the blue collars. Furthermore, it allows dynamic coordination of production orders for industrial plants via appropriate digital twin (DT) technologies. This paper proposes an original framework for RTLS in industrial environments and presents a case study for framework application at the TU Graz Learning Factory.
BibTeX
@article{wolf2022real, title={Real time locating systems for human centered production planning and monitoring}, author={Wolf, Matthias and Rantschl, Marvin and Auberger, E and Preising, Heimo and Sbaragli, Andrea and Pilati, Francesco and Ramsauer, Christian}, journal={IFAC-PapersOnLine}, volume={55}, number={2}, pages={366--371}, year={2022}, publisher={Elsevier} } - F. Pilati, A. Sbaragli, M. Nardello, L. Santoro, D. Fontanelli, D. Brunelli ‘Indoor positioning systems to prevent the COVID19 transmission in manufacturing environments’, Procedia CIRP, vol. 107, pp. 1588–1593, 2022. doi: .
Abstract
Since the 11th of March 2020 when the World Health Organization declared the novel COVID-19 outbreak a global pandemic, it registered officially over 5 million deaths worldwide. According to the course of the pandemic, governments encouraged best practices and then ruled out temporary restrictions on daily lives. In this scenario, non-essential labor-intensive sectors were forced to put on hold operations producing massive temporary layoffs. In gradually restoring the economic activities, governments passed several laws to passively mitigate the pathogen transmission in indoor working environments. However, several COVID19-related injuries were filled by manufacturing companies. According to the outlined conditions, this paper proposes an original and advanced hardware and software architecture to prevent the COVID19 transmission in indoor production environments. The aim is to increase the safety of whichever indoor productive workplace through a contact tracing approach. Indoor positioning systems due to their ability to accurately track the movement of tagged entities compose the hardware part. For this purpose, human operatives are equipped with adequate wearable sensors. Raw data acquired are properly mined through advanced algorithms to quantitatively assess the degree of safety of any working setting. Indeed, having as a reference the epidemiological evidence the software part defines an innovative risk index along two correlated dimensions. While the first defines the risk of any worker getting infected during the shift, the other one expresses the degree of COVID19-safety of the shop floor defined by the displacements of the anchors. Benefitting from these targeted and quantitative hints, plant supervisors may redesign the production settings to lower the chances of COVID19 infection. This innovative digital framework is validated in a real case study in the North of Italy which performs manual mechanical processing for the automotive industry.
BibTeX
@article{pilati2022indoor, title={Indoor positioning systems to prevent the COVID19 transmission in manufacturing environments}, author={Pilati, Francesco and Sbaragli, Andrea and Nardello, Matteo and Santoro, Luca and Fontanelli, Daniele and Brunelli, Davide}, journal={Procedia CIRP}, volume={107}, pages={1588--1593}, year={2022}, publisher={Elsevier} } - A. Sbaragli, F. Pilati, A. Regattieri, Y. Cohen, Others ‘Real time locating system for a learning cross-docking warehouse’, Social Sciences Research Network, pp. 1–6, 2021. doi: .
Abstract
Real time locating systems (RTLS) represent today an established and reliable technology to identify, track and monitor, in real time with any human commitment, the dynamic evolution of the spatial location of tagged entities inside factory layout. This results in more effective and continuous monitoring of products, stock-keeping units (SKU), and vehicles of production plants or logistic facilities. This paper targets the adoption of such RTLS in warehousing systems focusing on the related learning opportunities to enhance the efficacy and the productivity of the storage process which have consistent externalities during the order picking. For this purpose, a crossdocking warehouse is equipped with an ultrawideband (UWB)-based RTLS to track any forklifts traveling activities and SKU dynamic locations. A set of relevant data is automatically generated by the interactions between the transmitters, which are installed on board of the forklifts and connected to the SKU barcode readers, and the receivers, which are displaced in fixed positions all over the warehouse layout for optimizing the data transmissions. Furthermore, to yield further insights into the performed storage and retrieval operations, the daily incoming and outcoming shipping orders are also considered at the detail level of every single SKU. All this relevant information is merged into a unique database that evolves in real time representing the dynamic evolution of the warehouse operations over time. A digital representation of the physical storage system is developed to leverage such relevant datasets through adequate data analysis algorithms. A set of quantitative key performance indicators is evaluated, and suggestions are automatically offered to the warehouse manager to improve the storage system efficiency. The implementation of such modification, through a feedback loop, is to offer the unique opportunity to such warehousing system to analytically learn by its process’s underperformances, as well as to evaluate the efficacy and efficiency of the corrective actions suggested by the developed algorithms.
BibTeX
@article{sbaragli2021real, title={Real time locating system for a learning cross-docking warehouse}, author={Sbaragli, Andrea and Pilati, Francesco and Regattieri, Alberto and Cohen, Yuval and others}, journal={Social Sciences Research Network}, pages={1--6}, year={2021}, publisher={Institute for Innovation and Industrial Management} }
PhD Thesis
- Andrea Sbaragli ‘Cyber-physical systems to monitor the efficiency and sustainability of human-centric manufacturing systems’, Università degli studi di Trento, 2025.
Abstract
The manufacturing domain has been experiencing several revolutions over the years that have been shaping not only the design and management of processes but also their core drivers and value propositions. Industry 4.0 unleashes many enabling technologies such as the Internet of Things sensors and machine learning algorithms to boost industries’ productivity through data-driven process monitoring, rather than relying on operation manager experience. However, this fourth revolution does not set as strategic goals sustainability drivers (e.g., social and environmental) triggered by external forces that undermine modern societies. European policymakers address this structural limitation by defining the Industry 5.0 paradigm focused on human-centric and sustainable value creations. In this fast-paced landscape, this doctoral thesis targets the limitations of Industry 4.0 related contributions and defines three research questions to demonstrate the competitive advantages in designing cyber-physical systems to monitor the efficiency and sustainability of human-centric manufacturing environments. The human-centricity is an important feature of this work because, despite the rise of automation, workers represent a strategic and fragile resource in industrial plants. Therefore, Internet of Things technologies are leveraged to achieve a digital representation of workers. The acquired measurements are fed into computational algorithms to appreciate data-driven managerial insights based on the returned Key Performance and Risk indicators. The contributions of this thesis can be conceptually divided into two separate streams. The first demonstrates the relevance of enhancing the operational visibility of in-plant operations by exploiting Real Time Locating Systems acquisition layers. Although this technology indoor locates whichever (manufacturing) entity and asset in a defined coverage area, the returned workers’ positions fail to evaluate systems’ performances and sustainability. For this purpose, density-based machine learning algorithms and neural networks are introduced and validated to embed operational metrics into Decision Support Systems. Multidimensional managerial insights prove the consistency of this methodology in three different manufacturing environments. Considering production settings, managers appreciate the uptimes of workers and resource utilizations while evaluating the layout configurations and the related efficiency in manual material handling activities. This twofold level of analysis enables to eventually increase in-plant productivity while optimizing workers’ efforts in replenishment routes. The logistic investigation offers similar takes by monitoring the Overall Equipment Effectivness of manual forklifts and the distribution of picking/depositing activities in storage areas. Potential inefficiencies provide valid input to optimize the performances while reducing the energy consumption of logistics vehicles. The second stream focuses on workers’ physical resilience during task executions. To achieve this purpose, ergonomic indices are largely adopted to mitigate work-related musculoskeletal disorders in the workforce. The European Assembly Worksheet screening tool is the most complete one focusing on several parameters ranging from working postures to exterted forces. The developed cyber-physical system mirrors in digital spaces workers’ operations through a multi-device acquisition layer. While a four-channel surface ElectroMyoGraphy and a network of markerless cameras acquire muscular contractions in upper limbs and body joints, a radio-frequency-based smart glove detects process interactions such as tool usages and component pickings and thus segments production activities. These digital measurements are fed into computational algorithms to automate the mentioned ergonomic assessment. The experimental campaign validates the proposed cyber-physical systems and draws several managerial insights. For instance, strong bending postures may highlight a poor workplace design suggesting the need of self-adjustable workstations to accommodate a diverse workforce. At the same time, worrisome exerted forces could require line rebalancing to fairly redistribute muscular activity rates among operators. In summary, this thesis represents a significant advancement in digital manufacturing, offering ready-to-deploy systems while outlining future research opportunities and applications.
BibTeX
@article{sbaragli2025cyber, title={Cyber-physical systems to monitor the efficiency and sustainability of human-centric manufacturing systems}, author={Sbaragli, Andrea and others}, year={2025}, publisher={Università degli studi di Trento} }