A framework for data-driven digital twins of smart manufacturing systems

Sediment transport is a fundamental process shaping landscapes and posing significant hazards in mountainous regions. However, traditional field monitoring and simulation approaches, such as grain size sampling and numerical modeling, are often costly and time consuming. Recent advances in physics-based models and machine learning have substantially improved spatial and temporal resolution. These achievements enable the development of digital twins to explore what-if scenarios and to better understand the dynamic processes involved.In this work, we combine the probabilistic sediment cascade model (SedCas) with the machine learning–based event detection model (Flow-Alert) to develop a digital twin of a catchment. The former relies solely on climate forcing to simulate sediment dynamics, whereas the latter uses seismic signals to identify extreme sediment transport events, such as debris flows. We address three key questions. First, how to design a digital twin framework that captures the physical components of sediment transport, including erosion on hillslopes, hillslope to channel transfer, and channel transport to the catchment outlet, at hourly and even sub hourly temporal resolution. Second, how to fuse predictions from the physics-based model SedCas and the machine learning based model Flow-Alert to merge and balance the strengths of these two modeling approaches. Third, how to reduce uncertainty when translating insights from the virtual entity back to the physical entity. We demonstrate that the digital twin framework enables potential users, such as governmental agencies and local stakeholders, to explore what if scenarios and better understand how climate change and human interventions influence sediment transport dynamics.

The concept of Industry 4.0 uses cyber-physical systems and the Internet of Things to create "smart factories" that enable automated and connected production. However, the complex and diverse nature of cyber-physical systems makes them vulnerable to physical threats, where any fault in their physical components can lead to serious consequences, from financial losses to human safety. To reduce the risk of such failures, it is essential to have techniques in place to monitor the condition of these physical components closely. In recent years, data-driven predictive maintenance methods have become common due to the growth in data and advancements in machine learning. This thesis focuses on fault diagnosis, a key part of predictive maintenance, which aims to detect and identify faults in physical components. Incorporating advanced techniques, including machine learning algorithms and sensor data analysis, enhanced fault diagnosis systems' accuracy and efficiency. However, several real-world barriers challenge the effectiveness of these techniques. This research is structured around three primary objectives. The first objective aims to establish an efficient fault diagnosis system that is highly sensitive to fault occurrences while being resilient to uncertainties such as variability in fault patterns and data imbalance. To address variability in fault patterns, we developed a multi-step approach integrating clustering techniques and an adaptive neuro-fuzzy inference system, which significantly improved diagnosis robustness even under noisy conditions. For imbalanced datasets, a two-step method was introduced, combining permutation entropy for fault detection and envelope analysis for fault isolation. This approach was enhanced by an automatic frequency band selection method leveraging wavelet transform and statistical criteria, resulting in high fault diagnosis accuracy and reliability, even with limited labeled faulty data. The second objective focuses on developing fault diagnosis systems that can maintain high performance despite domain shifts and limited training data in both source and target domains. To address the challenge of domain shifts, a transfer learning technique was introduced, and a systematic literature review was conducted. This study bridges a technical gap by providing a comprehensive taxonomy, state-of-the-art methods, and insights into the challenges of applying transfer learning to predictive maintenance. Additionally, a conceptual framework was proposed to explore scenarios for transferring diagnostic knowledge from different source domains to target domains with insufficient training datasets. In a subsequent study, a new diagnostic framework was developed using digital twins and partial transfer learning to address the lack of real training datasets in both source and target domains as well as label space inconsistencies between them. The framework includes a general approach for developing digital twins of real machines to generate synthetic diagnostic data as source domain data. Furthermore, by incorporating a deep learning network and a double-layer attention mechanism, a partial transfer learning diagnostic model was developed to transfer diagnostic knowledge from the digital twin to the real target domain despite label space inconsistencies. The proposed framework was validated through an industrial case study involving rolling bearings in rotating equipment. The third objective focuses on developing an efficient fault diagnosis system capable of maintaining reliability under unseen and evolving working conditions. To achieve this, a novel self-adaptive fault diagnosis system was proposed, incorporating the MAPE-K loop, a domain generalization network model, and digital twin technologies. This self-adaptive fault diagnosis system leverages the MAPE-K loop to adapt in real time to evolving working conditions. Digital twins were employed to continuously generate data aligned with the real system’s operational context, enabling accurate fault diagnosis in dynamic environments. The domain generalization network model enhances generalization to unseen target working conditions by addressing overfitting and feature loss through multi-domain data augmentation, adversarial learning, and domain discrepancy modules. Validation using multiple rotating machinery datasets demonstrated that the developed fault diagnosis system outperformed state-of-the-art methods across various scenarios. In summary, this dissertation proposes a comprehensive and practical approach to addressing variability in fault patterns, imbalance datasets, data scarcity, domain shifts, and unseen and evolving working conditions challenges in existing fault diagnosis systems.

Monitoring manufacturing systems using AI: A method based on a digital factory twin to train CNNs on synthetic data

The rapid expansion of industrialization and the increase in energy demand make energy resource management necessary in various industries. Creating digital twins from available streaming data offers a practical way to enhance the efficiency of manufacturing systems. Digital twins allow for detailed analysis of key metrics, such as the energy consumption of complete manufacturing systems. Utilizing digital twins can greatly assist in improving manufacturing processes for reduced energy consumption. In this paper, we investigate the data requirements for the generation of energy-oriented digital twins of manufacturing systems. For this purpose, we developed a case study to illustrate and outline the necessary data requirements for a manufacturing system to enable the collection of data that can be used to extract energy-oriented simulation models. We, subsequently, analyze and generalize our findings that can be applied to a broader context.

As the Internet of Things (IoT) revolutionizes intelligent environments, energy efficiency remains a critical challenge. Traditional IoT-based energy management systems rely on predefined rule-based automation, which lacks adaptability to dynamic environmental conditions. This study proposes an Agentic AIoT system that integrates AI Agents, Large Language Models (LLMs), and Digital Twin technology to enhance IoT decision-making and optimize energy consumption in smart homes. Unlike conventional IoT frameworks, our system dynamically analyzes contextual factors—including real-time environmental conditions, user behaviors, and external data sources (e.g., weather forecasts and occupancy schedules)—to generate optimal energy-saving decisions autonomously. The proposed system consists of three key components: (1) Information Modules, which aggregate multi-source data beyond traditional IoT sensor inputs and construct a Digital Twin model for real-time environment simulation; (2) AI Agents & LLMs, which process contextual data and generate adaptive energy-saving strategies through intelligent reasoning; and (3) Automated Decision Execution, which seamlessly integrates optimized actions into IoT operations. By incorporating Digital Twin technology, the system can simulate various energy consumption scenarios, predict future power demands, and refine decision-making through AI-driven iterative learning. Experimental results demonstrate that our AIoT system achieves a 47.77% reduction in energy consumption compared to non-IoT environments while enhancing automation, adaptability, and user-centric control. Unlike conventional rule-based IoT solutions, integrating Digital Twin with LLM-powered AI Agents enables real-time context-aware decision-making and predictive energy optimization. This study contributes to intelligent IoT applications, Digital Twin-driven energy management, and AI-enhanced automation. Future research will focus on refining AI-driven autonomous decision-making, improving Digital Twin’s predictive accuracy, and further quantifying the impact of LLM-powered AIoT systems on energy efficiency. Our findings highlight the trans-formative potential of LLM-integrated AIoT with Digital Twin technology in revolutionizing energy management for intelligent environments while ensuring seamless interoperability with existing IoT infrastructures.

Biopharma Is Going Digital … Bit by Bit

Digital equivalence is the main objective of Digital Twin (DT)s, and simulation is an integral part. DTs reach beyond traditional simulation with the help of real-time synchronization through Industrial Internet of Things (IIoT) technologies. Simulation supports off-line experimentations and planning, while DTs offer synchronous execution and modification. DTs help to understand “what may happen”. Also, they present “what is happening” and its management methodologies. In this paper, we present building aspects and an integration approach for a Digital Twin based Discrete-Event Simulation model. Our approach utilizes a data-driven agent-based simulation within a DT framework. It presents an integration layer that provides two essential features: reconfiguration and state initialization. It gives simulation models configurability and integrity that are required for operating within a DT. Our proposed approach is presented through a use case of a semiconductor manufacturing system. The proposed integration layer extends the usability of current Discrete-Event Simulation (DES) for a DT within a Cyber-Physical Production System.

Digital Twin (DT) technology is revolutionizing industry in the Industry 4.0 era by enabling simulating, analyzing, and monitoring physical assets real-time. To reduce downtime for machines, improve maintenance schedules, and maximize equipment lifespan, this research explores the use of digital twins in implementing predictive maintenance (PdM) concepts. An imitation of an actual machine or process that is continuously supplied with information from embedded Internet of Things sensors is referred to as a "digital twin." Condition-driven maintenance, in contrast to reactive or time-based maintenance, is enabled by the system to predict failures before they occur by the integration of AI/ML algorithms. With reference to four key components—data collection, model building, simulation environment, and feedback mechanisms—the research provides a conceptual framework for the application of digital twins in a typical manufacturing environment. It also illustrates a case study of a CNC machine that employs real-time temperature and vibration data to anticipate spindle bearing issues. Compared to standard methods, DT-based predictive maintenance enhances asset availability, cost savings, and accuracy in maintenance significantly. To deliver insights into future directions and scalability of DT in smart factories, the article concludes by considering challenges such as data security, interoperability, and computing complexity. Keywords: Industry 4.0, IoT, Smart Manufacturing, Machine Learning, CNC, Condition Monitoring, Digital Twin, Predictive Maintenance.

Digital Twin-based bottleneck prediction for improved production control

The Digital Twin (DT) technology is considered as a backbone in the Industrial 4.0 revolution as it is playing a vital role in the digitization of various industries. A DT is a virtual representation of a physical entity, thus having the ability to simulate real data generated at physical space to optimize, estimate, control, monitor and forecast states / configurations. Despite enormous benefits, DT technology has several implementation challenges. Although deploying DT on edge or cloud platforms yields a plethora of services, its implementation in both spaces faces certain limitations. These limitations include latency, data communication overload, transmission energy consumption, privacy concerns, and communication inefficiencies. It is evident that these shortcomings could significantly impact real-time monitoring and control. Therefore, when considering whether to deploy DT on the edge or on the cloud, it is necessary to make a trade-off, or alternatively, adopt a hybrid approach. However, it is important to acknowledge that even with a hybrid approach, the aforementioned issues will persist to some extent. To address these challenges, this article introduces two innovative approaches. Local DT (LDT) and Distributed DT (DDT). These deployment strategies are designed to mitigate latency, minimize data communication overload, reduce energy consumption, improve communication efficiency, and strengthen privacy measures.Thus, resulting in environmental and economic sustainability. Consequently, these advancements facilitate superior real-time monitoring and control capabilities. Through the utilization of LDT and DDT methodologies, organizations can harness the full potential of DT technology, thereby maximizing its benefits.

The production of food is highly complex due to the various chemo-physical and biological processes that must be controlled for transforming ingredients into final products. Further, production processes must be adapted to the variability of the ingredients, e.g., due to seasonal fluctuations of raw material quality. Digital twins are known from Industry 4.0 as a method to model, simulate, and optimize processes. In this vision paper, we describe the concept of a digital food twin. Due to the variability of the raw materials, such a digital twin has to take into account not only the processing steps but also the chemical, physical, or microbiological properties that change the food independently from the processing. We propose a hybrid modeling approach, which integrates the traditional approach of food process modeling and simulation of the bio-chemical and physical properties with a data-driven approach based on the application of machine learning. This work presents a conceptual framework for our digital twin concept based on explainable artificial intelligence and wearable technology. We discuss the potential in four case studies and derive open research challenges.

Digital twins (DTs) are increasingly utilized to monitor, manage, and optimize complex systems across various domains, including civil engineering. A core requirement for an effective DT is to act as a fast, accurate, and maintainable surrogate of its physical counterpart, the physical twin (PT). To this end, machine learning (ML) is frequently employed to (i) construct real-time DT prototypes using efficient reducedorder models (ROMs) derived from high-fidelity simulations of the PT’s nominal behavior, and (ii) specialize these prototypes into DT instances by leveraging historical sensor data from the target PT. Despite the broad applicability of ML, its use in DT engineering remains largely ad hoc. Indeed, while conventional ML pipelines often train a single model for a specific task, DTs typically require multiple, task- and domain-dependent models. Thus, a more structured approach is required to design DTs. In this paper, we introduce DesCartes Builder, an opensource tool to enable the systematic engineering of ML-based pipelines for real-time DT prototypes and DT instances. The tool leverages an open and flexible visual data flow paradigm to facilitate the specification, composition, and reuse of ML models. It also integrates a library of parameterizable core operations and ML algorithms tailored for DT design. We demonstrate the effectiveness and usability of DesCartes Builder through a civil engineering use case involving the design of a real-time DT prototype to predict the plastic strain of a structure.

Digital twin technology has been increasingly applied in healthcare and patient well-being in recent years. This paper provides an overview of the current methods and applications of digital twins in the healthcare field. One such application is digital twins in precision healthcare, where digital twins are used to create patient-specific models to assist in diagnosis and treatment planning. Digital twins are also used in hospital/clinic management, where they help to optimize resource allocation and workflow processes. In response to the COVID-19 pandemic, digital twins have been utilized to detect outbreaks and predict disease spread. In addition, digital twins have been applied in bio-manufacturing and pharmaceutical industry to improve manufacturing processes. Another application area is machine learning and modeling, where digital twins are used in machine learning, data generation, and system modeling for applications in healthcare and disease prediction. Security and ethical issues related to digital twins are also discussed in this paper, as privacy concerns and data protection remain important considerations in the application of digital twin technology in healthcare. Finally, the paper concludes by discussing the future challenges and directions of future work in this field. These include the need to develop more accurate and sophisticated digital twin models, addressing interoperability and integration issues, and further exploring the potential of digital twin technology in emerging areas such as telemedicine and personalized medicine.

In modern medicine, novel technologies play a key role in improving the diagnosis, treatment, and management of patient health. One of the most promising innovations is Digital Twin technology, which enables the creation of virtual models of real-world objects, including human organs. This technology opens new possibilities for personalised medicine, disease prediction, and the optimisation of medical processes. Digital Twins are already actively used in various industries, such as manufacturing, aviation, and urban planning, but their potential in medicine is only beginning to unfold. They allow for creating detailed models of an individual patient’s body, contributing to more accurate diagnostics, simulation of treatment responses, and enhanced efficiency of medical interventions. This article explores the key aspects of Digital Twin medical applications, their advantages, challenges, and development prospects. Special attention is given to the technical foundations of this technology, its use in personalised medicine, health monitoring, disease prediction, and the optimisation of healthcare facilities. For the practical part, the study shows how to create a Digital Twin of a heart based on real patient data. Additionally, the study investigates a model of the heart’s Digital Twin, demonstrates its use, and illustrates how to replicate heart functionality using the Digital Twin. The study describes how a part of a Digital Twin was created using the suggested architecture and outlines how to use and extend this functionality. More than that, the study shows how to integrate artificial intelligence into the Digital Twin architecture and how artificial intelligence algorithms help create Digital Twins. The research shows how to build the Digital Twin using machine learning and artificial intelligence technologies, and how to choose the mathematical model for the Digital Twin. To conclude, the study defines further work, problems, and difficulties in creating the Digital Twin of the heart. The study shows the perspectives on using Digital Twins, their practicality, and real-life cases where Digital Twins can save lives. Also, the study defines the next steps for the Digital Twin creation, testing, implementation, and usage in the real-life healthcare industry.

Computational simulation approaches have long been a cornerstone of design in marine and offshore structures. More recently, they have also become essential for optimizing operations and ensuring sustainability. The digital twin (DT) framework has enabled to bridge the gap between simulation and its practical application in controlling, monitoring, and operating marine assets. However, the applications where complex non-linear problems are solved in real or close to real-time are still limited. It is proposed that this could be overcome by the recent rise of artificial intelligence (AI) and machine learning methods, which are promising technologies that can make current computational models more versatile. Namely, data-driven surrogate models or neural networks embedded in numerical algorithms enable the relegation of computationally complex calculations into simple, functional forms, opening a pathway to a new generation of digital twins. Memory-augmented neural networks offer a viable solution for applications involving memory-dependent problems, such as ship motion control or material fracture. These neural network models operate incrementally, much the same as the computational models they are inserted to. This work explores a critical yet underexamined aspect of these incremental machine learning models that is important when considering the pathway to real-time digital twins for ship motion control or non-linear structural analysis.