In modern industries, the integration of system- and system-of-systems (SoS) digital twins (DTs) emerged as a powerful tool for performance monitoring, deviation analysis and predictive maintenance. This study demonstrates how these DTs can assist in troubleshooting the complex multi-layer process for identifying process related issues and failure. This study also highlights experience in implementing DTs in industrial environments, such as the complexity of processes, managing and exchanging real-time data between the manufacturing processes and analytical systems. The research includes a case study on an industrial ring rolling process, where multiple DTs were deployed in-parallel with the operation to continuously monitor the entire process, including individual systems and their interconnections. System-level DTs were combined to construct SoS-level DT, enabling comprehensive performance evaluation. The analytical system was developed by using Python modules, monitoring real-time data over a period of ten months. To aid industrial users, high-dimensional analyses from both system- and SoS-level DTs were synthesized using Principal Component Analysis, providing a quick overview of the overall process. The results illustrated the problem that appeared in the rolling process were due to one of its processing tools. The process anomalies at early stage, assisting in identifying the root causes were also highlighted in this study. Furthermore, there were four challenges experienced in the research, i.e., monitoring and troubleshooting complexity, computing and managing real-time data, handling false alarm, and processing data close to edge devices.

Anomaly Detection in Argon Oxygen Decarburization (AOD) Processes is crucial for optimal functioning and for uninterrupted production in steel industry. In this prestudy, we aim to detect an anomaly called calibration error on a flow sensor in AOD converter. To this purpose, we acquire data from Alleima, a steel company in Sweden. This data consists of signals from five different sensors including two flow sensors for measuring oxygen and argon flows, control signals for the corresponding valves and the pressure after the gases are mixed. This data is manually annotated as calibration error on a flow sensor-based anomaly (CEFSBA) and normal class by the experts. From each of the sensor signals, two features namely entropy and variance are extracted. These features are fed to classifier to classify a feature vector into normal and CEFSBA. Our proposed approach detected CEFSBA with an accuracy of 92.1% when using neural networks. Our prestudy, indicates that machine learning-based approaches can be used to identify CEFSBA in the AOD converter. However, in order to use in industry, the performance needs to be improved before deploying it for real time CEFSBA detection.

Accurately identifying moisture content in wood chips is crucial for optimizing energy production. This paper presents an automated classification approach using ultrawideband (UWB) radio transmission data. We collect data from 1,923 samples across four power plants and extract seven key features based on zero-crossings and amplitude information. To enhance classification performance, we apply Chi-square feature selection to identify the most significant features. These selected features are then fed into a classifier to determine moisture content levels. Our approach achieves a classification accuracy of 85.26%, demonstrating the effectiveness of the extracted features and the proposed methodology.

In controlling the State of Health (SOH) of batteries, understanding the individual and collective impact of physical factors such as current draw, temperature, and age is crucial. Traditional lab tests offer deep insights into battery degradation but cannot isolate the effects of specific variables, as they typically analyze a series of actions and conditions in combination. Although this is useful for estimating the SOH, it limits the ability to develop strategies for optimal control of the battery lifetime. To address this, as a first step, this article proposes a method for designing a function estimator based on the State–Action Cost (SAC) that considers current, voltage, temperature, and age to predict the micro-effects of each state– action combination on the SOH. This estimator will provide a more granular understanding of how individual actions and conditions contribute to battery health. In the second step, it will be demonstrated how this approach could significantly enhance the precision of SOH control and facilitate better management strategies for extending battery lifespan.

This paper presents an integrated approach combiningprescriptive maintenance with adaptive production planning,leveraging Digital Twin technology to optimize asset management.By adjusting production schedules based on real-time healthmonitoring, the proposed method aims to improve maintenanceefficiency, reduce downtime, and enhance overall productivity.The integration of predictive models with optimization techniquesallows for more informed decision-making, ensuring that maintenanceinterventions are timely and cost-effective while minimizingthe risk of system failure. This approach offers significantpotential for industrial settings, where both operational efficiencyand equipment longevity are critical.

Controlling the machine’s state of health (SoH) increases the accuracy of the remaining useful life estimation and enables the control of the failure time by keeping the system operational until the desired maintenance time is reached. To achieve system reliability through SoH control, the system controller must consider the impact of its actions on other parameters, such as degradation. This article proposes a structure for designing degradation-aware controllers for systems with available physical models. A system using this approach can learn autonomously, irrespective of the system’s physical structure and degradation model, and opt for control actions that enhance the system’s reliability and availability. To this end, first, a method is proposed to compute the cost associated with the actions taken by the controller. Second, a new cost function is introduced that incorporates the costs associated with degradation into the cost function utilized in model predictive control. In the third step, dynamic programming and deterministic scheduling are used to calculate the optimal action based on the defined cost function. Finally, the proposed control method is validated through simulation, demonstrating its ability to effectively manage machine degradation and achieve optimal performance according to production and maintenance plans.

This study aims to investigate the deployment and maintenance of Digital Twin (DT) in an industrial ring rolling manufacturing process. Three parameters were analyzed (i) data cleaning and processing, (ii) DT development and updating, and (iii) secure communication. A process-based DT model was developed using iba-ag and Python, prioritizing security throughout the process. The study underscores the significance of secure DT system, with particular attention to the challenges of real-time updates affected by process operations and environmental factors. This emphasizes the adaptability of the DT while acknowledging the expertise of process operators. The findings promote a deeper understanding of DT implementation, particularly for process developers, maintenance personnel, and operators. The research highlights the importance of security in DT systems, especially when dealing with industrial data influenced by production and environmental factors. Therefore, by focusing on security and deployment in an open-source environment, this study contributes to the practical use of DT systems for predictive maintenance and process optimization within the industrial settings.

This study introduces a comprehensive framework designed to enhance production efficiency by integrating maintenance strategies, energy costs, and production specifications. This integration is achieved through a novel empirical method for estimating state–action costs, suitable for both machines with measurable and non-measurable states-of-health. We address the challenge of under-determination in state–action cost optimization by employing a k-means clustering approach, ensuring robustness and applicability. Utilizing an adapted SARSA algorithm, our framework optimally controls shop-floor machinery to minimize the global cost function. The efficacy of the state–action cost estimation method is validated using NASA’s C-MAPSS dataset. Additionally, the optimization strategy is further corroborated through its successful implementation in an autonomous mining cart model on the shop floor. Our results highlight the framework’s ability to optimize machine lifetime and production processes effectively, providing tailored solutions that adapt to varying operational conditions without depending on predefined machine degradation models and costs.

Construction and manufacturing workers undertake physically laborious activities which put them at risk of developing serious musculoskeletal disorders (MSDs). In the EU, millions of workers are being affected by workplace-related MSDs, inflicting huge financial implications on the European economy. Besides that, increased health problems and financial losses, severe shortages of skilled labor also emerge. The work aims to create awareness and accelerate the adoption of exoskeletons among SMEs and construction workers to reduce MSDs. Large-scale manufacturers and automobile assemblers are more open to adopt exoskeletons, however, the use of exoskeletons in small and medium enterprises (SMEs) is still not recognized. This paper presents an experimental study demonstrating the advantages of different exoskeletons while performing workers’ tasks. The study illustrates how the use of certain upper and lower body exoskeletons can reduce muscle effort. The muscle activity of the participants was measured using EMG sensors and was compared while performing designated tasks. It was found that up to 60% reduction in human effort can be achieved while performing the same tasks using exoskeletons. This can also help ill workers in rehabilitation and putting them back to work. The study concludes with pragmatic recommendations for future exoskeletons.

The capability of estimating future maintenance needs in advance and in a timely manner is a prerequisite for reliable manufacturing with high availability in a production unit. Additionally, conducting planned maintenance efforts regularly and prematurely increases the service lifetimes and utilization rates of parts, which leads to more sustainable production. The benefits of predictive maintenance are obvious, but introducing it into a facility poses various challenges. In this study, digital twins of well-functioning machines are used for predictive maintenance. The discrepancies between each physical unit and its digital twin are used to detect the maintenance needs. A thorough evaluation of the method over a period of 18 months by comparing digital twin detection results with maintenance and control system logs shows promising results. The method is successful in detecting discrepancies, and the paper describes the techniques that are used. However, not all discrepancies are related to the maintenance needs, and the evaluation identifies and discusses the most common sources of error. These are often the results of human interaction, such as parameter changes, maintenance activities and component replacement.