In recent years, intelligent fault diagnosis based on domain adaptation has been used to address domain shifts in cyber–physical systems; however, the need for acquiring target data sufficiently limits their applicability to unseen working conditions. To overcome such limitations, domain generalization techniques have been introduced to enhance the capacity of fault diagnostic models to operate under unseen working conditions. Nevertheless, existing approaches assume access to extensive labeled training data from various source domains, posing challenges in real-world engineering scenarios due to resource constraints. Moreover, the absence of a mechanism for updating diagnostic models over time calls for the exploration of self-adaptive generalized diagnosis models that are capable of autonomous reconfiguration in response to new unseen working conditions. In such a context, this paper proposes a self-adaptive fault diagnosis system that combines several paradigms, namely Monitor-Analyze-Plan-Execute over a shared Knowledge (MAPE-K), Domain Generalization Network Models (DGNMs), and Digital Twins (DT). The MAPE-K loop enables run-time adaptation to dynamic industrial environments without human intervention. To address the scarcity of labeled training data, digital twins are used to generate supplementary data and continuously tune parameters to reflect the dynamics of new unseen working conditions. DGNM incorporates adversarial learning and a domain-based discrepancy metric to enhance feature diversity and generalization. The introduction of multi-domain data augmentation enhances feature diversity and facilitates learning correlations among multiple domains, ultimately improving the generalization of feature representations. The proposed fault diagnosis system has been evaluated on three publicly available rotating machinery datasets to demonstrate its higher performance in cross-work operation and cross-machine tasks compared to other state-of-the-art methods.

In Industry 4.0, efficient fault diagnosis is crucial for predictive maintenance but is often hindered by significant domain shifts between training and testing domains and lack of training datasets, limiting the effectiveness of machine learning in practice. Transfer learning has been used to address these challenges by utilizing knowledge from similar source domains. However, the scarcity of faulty data from real machines and the difficulty of obtaining labeled datasets from lab machines as source domain pose significant obstacles. This paper presents a novel diagnostic framework that integrates digital twins and transfer learning to overcome these limitations. Digital twins generate training datasets, while a model update strategy based on parameter sensitivity analysis improves adaptability. The framework also incorporates a partial transfer diagnostic model with a double attention mechanism to handle data distribution discrepancies and label inconsistencies between digital twins and real machines. Validated on an industrial rotating machine case study using real data, the proposed approach improves diagnostic accuracy by over 10% compared to state-of-the-art methods.

In the last years, there has been a growing interest in the emerging concept of digital twin (DT) as it represents a promising paradigm to continuously monitor cyber-physical systems, as well as to test and validate predictability, safety, and reliability aspects. At the same time, artificial intelligence (AI) is exponentially affirming as an extremely powerful tool when it comes to modeling the behavior of physical assets allowing, de facto, the possibility of making predictions on their potential evolution. However, despite the fact that DTs and AI (and their combination) can act as game-changing technologies in different domains (including the railways), several challenges have to be faced to ensure their effectiveness, especially when dealing with safety-critical systems. This paper provides a narrative review of the scientific literature on DTs for railway maintenance applications, with a special focus on their relationship with AI. The aim is to discuss the opportunities the integration of these two technologies could open in railway maintenance applications (and beyond), while highlighting the main challenges that should be overcome for its effective implementation.

The advent of Industry 4.0 has resulted in the widespread usage of novel paradigms and digital technologies within industrial production and manufacturing systems. The objective of making industrial operations monitoring easier also implied the usage of more effective data-driven predictive maintenance approaches, including those based on machine learning. Although those approaches are becoming increasingly popular, most of the traditional machine learning and deep learning algorithms experience the following three major challenges: 1) lack of training data (especially faulty data), 2) incompatible computation power, and 3) discrepancy in data distribution. A new data-driven technique, such as transfer learning, can be developed to overcome the issues related to traditional machine learning and deep learning for predictive maintenance. Motivated by the recent big interest towards transfer learning within computer science and artificial intelligence, in this paper we provide a systematic literature review addressing related research with a focus on predictive maintenance. The review aims to define transfer learning in the context of predictive maintenance by introducing a specific taxonomy based on relevant perspectives. We also discuss current advances, challenges, open-source datasets, and future directions of transfer learning applications in predictive maintenance from both theoretical and practical viewpoints.

Digital Twin is an emerging technology that is used for different purposes, e.g monitoring, optimization, prediction, etc., in a wide range of real-world applications. Manufacturing is the most prevalent industry employing digital twin technology to achieve sustainability through enhancing smartness and intelligence. In this regard, several literature reviews have been established on the digital twin's role in sustainable manufacturing development. However, despite the importance of assessment and evaluation of developed sustainable actions and policies, and the high capability of the digital twin concept to support it, there is a lack of effort to systematically review the current state-of-the-art on the contribution of the digital twin in sustainability assessment and policy evaluation. By conducting a systematic literature review, this paper seeks to close this gap. By applying inclusion and exclusion criteria, 12 relevant papers are identified to be analyzed in more detail. The results show the ongoing effort on developing architectural frameworks and cutting-edge methodologies for integrating Digital Twin with conventional sustainability assessment and policy evaluation approaches. However, its potential benefits are not fully utilized, as evidenced by the limited effort put forth in this direction.

In the last years, there has been a growing interest in the emerging concept of digital twins (DTs) among software engineers and researchers. DTs not only represent a promising paradigm to improve product quality and optimize production processes, but they also may help enhance the predictability and resilience of cyber-physical systems operating in critical contexts. In this work, we investigate the adoption of DTs in the railway sector, focusing in particular on the role of artificial intelligence (AI) technologies as key enablers for building added-value services and applications related to smart decision-making. In this paper, in particular, we address predictive maintenance which represents one of the most promising services benefiting from the combination of DT and AI. To cope with the lack of mature DT development methodologies and standardized frameworks, we detail a workflow for DT design and development specifically tailored to a predictive maintenance scenario and propose a high-level architecture for AI-enabled DTs supporting such workflow.

Rotating equipment is considered as a key component in several industrial sectors. In fact, the continuous operation of many industrial machines such as sub-sea pumps and gas turbines relies on the correct performance of their rotating equipment. In order to reduce the probability of malfunctions in this equipment, condition monitoring, and fault diagnosis systems are essential. In this work, a novel approach is proposed to perform fault diagnosis in rotating equipment based on permutation entropy, signal processing, and artificial intelligence. To that aim, vibration signals are employed for an indication of bearing performance. In order to facilitate fault diagnosis, fault detection and isolation are performed in two separate steps. As first, once a vibration signal is received, the faulty state of the bearing is determined by permutation entropy. In case a faulty state is detected, the fault type is determined using an approach based on signal processing and artificial intelligence. Wavelet packet transform and envelope analysis of the vibration signals are utilized to extract the frequency components of the fault. The proposed approach allows for the automatic selection of a frequency band that includes the characteristic resonance frequency of the fault, which is subject to change in different operational conditions. The method works by extracting the proper features of the signals that are used to decide about the faulty bearing’s condition by a multi-output adaptive neuro-fuzzy inference system classifier. The effectiveness of the approach is assessed by the Case Western Reserve University dataset: the analysis demonstrates the proposed method’s capabilities in accurately diagnosing faults in rotating equipment as compared to existing approaches.

An essential aspect of resilience within Cyber-Physical Systems stands in their capacity of early detection of faults before they generate failures. Faults can be of any origin, either natural or intentional. Detection of faults enables predictive maintenance, where faults are managed through diagnosis and prognosis. In this paper we focus on intelligent predictive maintenance based on a class of machine learning techniques, namely transfer learning, which overcomes some limitations of traditional approaches in terms of availability of appropriate training datasets and discrepancy of data distribution. We provide a conceptual approach and a reference architecture supporting transfer learning within intelligent predictive maintenance applications for cyber-physical systems. The approach is based on the emerging paradigms of Industry 4.0, the industrial Internet of Things, and Digital Twins hosting run-time models for providing the training data set for the target domain. Although we mainly focus on health monitoring and prognostics of industrial machinery as a reference application, the general approach is suitable to both physical- and cyber-threat detection, and to any combination of them within the same system, or even in complex systems-of-systems such as critical infrastructures. We show how transfer learning can aid predictive maintenance with intelligent fault detection, diagnosis and prognosis, and describe some the challenges that need to be addressed for its effective adoption in real industrial applications.

In the last years, there has been a growing interest in the emerging concept of Digital Twins (DTs) among software engineers and researchers. DTs represent a promising paradigm to enhance the predictability, safety, and reliability of cyber-physical systems. They can play a key role in different domains, as it is also witnessed by several ongoing standardisation activities. However, several challenging issues have to be faced in order to effectively adopt DTs, in particular when dealing with critical systems. This work provides a review of the scientific literature on DTs in the railway sector, with a special focus on their relationship with Artificial Intelligence. Challenges and opportunities for the usage of DTs in railways have been identified, with interoperability being the most discussed challenge. One difficulty is to transmit operational data in real-time from edge systems to the cloud in order to achieve timely decision making. We also provide some guidelines to support the design of DTs with a focus on machine learning for railway maintenance. 

The Internet of Things (IoT) is rapidly developing in diverse and critical applications such as environmental sensing andindustrial control systems. IoT devices can be very heterogeneous in terms of hardware and software architectures, communication protocols, and/or manufacturers. Therefore, when those devices are connected together to build a complex system,detecting and fxing any anomalies can be very challenging. In this paper, we explore a relatively novel technique known asProcess Mining, which—in combination with log-fle analytics and machine learning—can support early diagnosis, prognosis, and subsequent automated repair to improve the resilience of IoT devices within possibly complex cyber-physicalsystems. Issues addressed in this paper include generation of consistent Event Logs and defnition of a roadmap towardefective Process Discovery and Conformance Checking to support Self-Healing in IoT.