The structural health monitoring (SHM) of large steel box girders often lacks baseline data, making traditional damage detection methods unsuitable for structures in long-term service. To overcome this problem, this study proposes a baseline-free frequency response function (FRF)-wavelet packet permutation entropy (WPPE)-local outlier factor (LOF) damage identification framework that integrates multi-source information fusion theory and sparse field inversion. Firstly, a composite damage indicator was constructed by integrating FRF, WPPE, and LOF, which can highlight non-stationary, frequency sensitive, and edge localization damage characteristics. In order to improve engineering interpretability and spatial robustness, an engineering prior weighting scheme based on stress distribution was introduced in the damage mapping stage, especially for the bending dominant region. Subsequently, a sparse field inversion method was developed by linking the indicators of the sensor domain with the stiffness attenuation at the unit level through weighted optimization. This makes the damage vector interpretable, thereby further deriving quantitative damage depth and severity. The proposed method was validated using a steel box girder finite element model and a triangular impact load induced broadband vibration test under healthy and three types of damage conditions. The results show that this method achieves accurate identification of damage locations, enhances sensitivity to slight and boundary damage, has strong robustness to noise and uncertainty of excitations, and does not require any baseline measurements. Due to these advantages, the proposed framework has great potential for application in large-span bridges or other large civil structures where baselines are difficult to obtain.

Ensuring reliability and safety is essential in complex energy systems such as wind turbines, where failures can trigger unexpected downtimes, severe incidents, and significant costs. This study proposes a hybrid BowTie-based reliability framework that integrates Fault Tree Analysis, Reliability Block Diagrams, and BowTie methodology to quantify risk and evaluate the effectiveness of safety barriers. The framework employs key reliability metrics including availability, probability of failure on demand, and probability of failure per hour, and supports scenario-based sensitivity analyses to explore redesign options. A simulation-based case study of a wind turbine generator subsystem is presented, using parameter values drawn from published reliability data. Results highlight that protective relays and automatic trip systems represent critical single points of defence, while improvements such as enhanced oil analysis and redundant dashboards reduce consequence frequency from 2.912 × 10−17 to 8.257 × 10−19 failures/h (a 97.16% reduction, nearly two orders of magnitude). Compared to conventional models, the proposed framework introduces explicit defence in depth modelling, improves computational compactness, and provides a practical decision support tool for asset managers by balancing safety and reliability. At this stage, the study should be regarded as a proof of concept that demonstrates feasibility and sets a foundation for future research and application to larger, more complex infrastructures.

Structural health monitoring (SHM) is essential for ensuring the safety and reliability of bridge structures. This study focuses on the application of Wavelet Packet Permutation Entropy (WPPE) for identifying and localizing multiple damage scenarios in steel box girder bridges under harmonic excitation. A finite element model of a steel box girder bridge was developed, and multiple damage cases with varying locations and orientations were simulated. WPPE was applied to extract damage-sensitive features from vibration responses, and its effectiveness in multi-damage identification was evaluated. The results demonstrate that WPPE can accurately detect and distinguish multiple damaged regions, even under complex conditions. The Permutation Entropy Difference (PED) values exhibit distinct peaks at multiple damage locations, enabling precise identification. While mutual interference between damage regions slightly affects detection accuracy, WPPE remains robust under 10 dB ambient noise, confirming its resistance to environmental interference. WPPE provides a sensitive and reliable approach for multi-damage identification, outperforming conventional frequency-domain methods in capturing nonlinear damage features. Future work will focus on experimental validation, application to large-scale bridge structures, and integration with deep learning models for enhanced automation and accuracy.

In vibration-based condition monitoring of rotating machinery, machine learning (ML) models have demonstrated significant diagnostic capabilities; however, their efficacy is fundamentally constrained by the selection and quality of input parameters. Current literature highlights a critical limitation: the absence of a unified parametric framework, with researchers consistently employing different vibration parameters and signal processing techniques for individual machine configurations. This inconsistency creates implementation challenges for industrial applications, where standardised methodologies are essential for reliable diagnostics across diverse mechanical systems. This study addresses this fundamental gap by proposing a standardised set of vibration parameters derived from poly-Coherent Composite Spectrum (pCCS) analysis that effectively captures fault signatures across different rotating machine configurations while significantly reducing computational overhead. The methodology integrates carefully selected time-domain parameters with frequency-domain parameters to comprehensively characterise rotor and bearing faults. Experimental validation is conducted on two distinctly different test rigs—a multi-rotor system operating across different speed regimes and a bearing configuration operating at three different speeds—representing completely different machine dynamics. Despite these substantial differences in mechanical configuration, the artificial neural network trained on these standardised parameters accurately classified multiple fault conditions (healthy states, misalignment, unbalance, shaft cracks, rotor-stator rub, and bearing faults) with near-perfect accuracy across all tested speeds and configurations. The results demonstrate that the proposed and properly selected vibration parameters based on rotordynamic principles can serve as standardised diagnostic features applicable to any rotating machine, offering a significant advancement toward unified condition monitoring frameworks for industrial implementation.

The present research combines machine learning approaches with poly-coherent composite spectrum (pCCS) analysis to propose a vibration-based fault detection solution for rotating machinery. Through a mathematical fusion of vibration measurements obtained from distributed bearing locations, the pCCS technique constructs a unified spectral signature, enabling systematic fault detection. Using pCCS substantially reduces the frequency domain parameters compared to analysing individual spectra from each measurement point. An artificial neural network (ANN) is used and trained on the extracted parameters for fault detection. The methodology is tested on an experimental rotating machine. This research investigates a range of machine states, from healthy conditions to experimentally simulated faults such as bearing defects, misalignment, shaft cracks, and rotor–stator rub. Integrating pCCS analysis with machine learning techniques aims to enhance the robustness, computational efficiency, and real-world application of defect detection in rotating machinery.

In rotating machines, any faults in anti-friction bearings occurring during operation can lead to failures that are unacceptable due to considerable downtime losses and maintenance costs. Hence, early fault detection is essential, and different vibration-based methods (VBMs) are explored to recognise incipient fault signatures. Based on rotordynamics, if a bearing defect causes metal-to-metal (MtM) impacts during shaft rotation, the impacts excite high-frequency resonance responses of the bearing assembly. The defect-related frequencies are modulated with the resonance responses and rely on signal demodulation for fault detection. However, the current study highlights that the bearing fault/faults may not be detected if the defect in a bearing is not causing MtM impacts nor exciting the high-frequency resonance of the bearing assembly. In a roller bearing, a localised defect may maintain persistent contact between rolling elements and raceways, thereby preventing the occurrence of impulse vibration responses. Due to contact persistence, such defects may not generate impact and may not be detected by existing VBMs, and the bearing could behave as healthy. This paper investigates such specific cases by exploring the relationship between roller-bearing defect characteristics and their potential to generate impact loads during operation. Using an experimental bearing rig, different roller and inner-race defects are presented while their fault characteristic frequencies remain undetected by the envelope analysis, fast Kurtogram, cyclic spectral coherence, and tensor decomposition methods. This study highlights the significance of both the dimension and location of defects within bearings on their detectability based on the rotordynamics concept. Further, simple roller-beam experiments are carried out to visualise and validate the reliability of the experimental observations made on the roller bearing dynamics.

In light of the susceptibility of steel box girder orthotropic plates to damage and cracking, the test data is utilized for structural damage identification. Vibration acceleration data is subjected to wavelet packet decomposition (WPD) and reconstruction, and permutation entropy (PE) is introduced to construct a damage index based on permutation entropy difference (PED). Through the non-destructive and damage dynamic test of the orthotropic steel bridge deck, the acceleration response signals of the components in the non-destructive and damaged states are respectively subjected to wavelet packet decomposition. Next, identify the component with the highest energy post-normalization for reconstruction. Subsequently, compute the permutation entropy of the reconstructed signal and ultimately perform damage identification by analyzing differences in permutation entropy. Additionally, a numerical simulation is conducted to establish the orthotropic steel bridge deck model, and the robustness of the wavelet packet permutation entropy to noise interference is analyzed. The research findings demonstrate that the wavelet packet permutation entropy-based structural damage identification method is highly sensitive to damages and robust to noise interference.

As a management tool, risk-based inspection (RBI) addresses an area of risk management not completely addressed in other organizational risk management Efforts, such as process hazard analysis (PHA) or reliability-centered maintenance (RCM). The RBI approach is a defined process for establishing and managing an inspection program based on understanding the failure probability and consequences of each equipment item. The RBI approach can focus the inspection program of the facility on the higher-risk equipment items, reducing the overall plant risk of a catastrophic failure while simultaneously providing a significant reduction to the cost of the ongoing inspection program. Moreover, it ensures all damage mechanisms identified in the corrosion study are being addressed. This paper outlines the different types of RBI models, i.e., qualitative, semi-quantitative, and quantitative models. Moreover, it provides insights into the basis of the most widely used quantitative RBI models in the oil and gas industry, i.e., API risk models, and then implemented them through a case study at an offshore gas production platform to evaluate and critically discuss the difference in the calculated consequences of failure resulting from the two methodologies of estimating the impact areas. The case study presented in this paper demonstrated the inconsistency in the calculated risk resulting from the two risk models, whereas the difference was several orders of magnitude for some equipment items. The resulting inspection and maintenance plans are likely to be significantly different if the same risk matrix and risk tolerance are used for both risk models.

Damage identification, both in terms of size and location, in bridges is important for timely maintenance and to avoid any catastrophic failure. An earlier experimental study showed that damage in a steel box girder orthotropic plate can be successfully detected using the measured vibration acceleration data. In this study, the wavelet packet decomposition (WPD) method is used to analyze the measured vibration acceleration responses and then the estimation of the permutation entropy (PE) on the re-constructed signals. A damage index is then defined based on the permutation entropy difference (PED) between the damaged and the healthy conditions to detect the location and size of the damage. The method is further validated through the finite element (FE) model of a steel box girder and the computed vibration acceleration responses when subjected to the sinusoidal excitations at different frequencies. In addition, the robustness of the methodology under different white noise interference conditions is also verified. The results show that the proposed methodology can effectively identify the location of human-made damage and accurately estimate the degree of damage under different frequencies of sinusoidal excitation. The method has shown a strong anti-noise property.

The 2-Steps Smart Rotor Fault Diagnosis Model (SRFDM) is proposed. This consists of a supervised classic pattern recognition artificial neural network (ANN), which uses parameters extracted from the measured vibration signals from the machine. The Step-1 identifies the machine is healthy or faulty, and then the classification of faults in the Step-2 is performed. Earlier studies have used both time and frequency domain parameters as the input vectors to the ANN model. Currently these parameters are normalised with the speed synchronous vibration amplitude from the frequency domain analysis to remove the influence of the machine unbalance due to change in the machine speeds. Hence, the proposed model is likely to be applied to a typical machine irrespective of the machine operating speeds.

Earlier studies have optimised the vibration-based parameters to identify the rotor defects only for the rotating machines. The artificial neural network (ANN) model was used earlier to classify the faults. The earlier optimised parameters are further examined for both rotor and bearing defects. These parameters are slightly modified in this research to accommodate bearing defects. The paper presents the study using an experimental vibration data from a laboratory-scaled rig.

Bearings are pivotal components of rotating machines where any defects could propagate and trigger systematic failures. Once faults are detected, accurately predicting remaining useful life (RUL) is essential for optimizing predictive maintenance. Although data-driven methods demonstrate promising performance in direct RUL prediction, their robustness and practicability need further improvement regarding physical interpretation and uncertainty quantification. This work leverages variational neural networks to model bearing degradation behind envelope spectra. A convolutional variational autoencoder for regression (CVAER) is developed to probabilistically predict RUL distributions with confidence measures. Enhanced average envelope spectra (AES) are used as network input for its physical robustness in bearing condition assessment and fault detection. The use of the envelope spectrum ensures that it contains only bearing-related information by removing other rotor-related frequencies, hence it improves the RUL prediction. Unlike traditional variational autoencoders, the probabilistic regressor and latent generator are formulated to quantify uncertainty in RUL estimates and learn meaningful latent representations conditioned on specific RUL. Experimental validations are conducted on vibration data collected using multiple accelerometers whose natural frequencies cover bearing resonance ranges to ensure fault detection reliability. Beyond conventional bearing diagnosis, envelope spectra are extended for statistical RUL prediction integrating physical knowledge of actual defect conditions. Comparative and ablation studies are conducted against benchmark models to demonstrate their effectiveness.

Condition-based monitoring (CBM) plays a fundamental role in bearing prognostics health management. To effectively reveal the degradation severity of the bearings, the choice of health indicators based on sensor measurements determines the promptness of fault detection and predictive maintenance. The Gini index (GI) is a widely used inequality measure for income distributions in economics while recent research has also successfully applied it in the CBM field. This study applies the Gini index of the vibration envelope signal (GIES) for bearing degradation detection and life prediction. Based on the changes of vibration impulsiveness, bearing degradation can be detected using the empirical three sigma rule. After the detection of bearing degradation, the remaining useful life (RUL) is predicted using the particle filter (PF) based on only a few GIES observations. The performance the RUL prediction method is demonstrated through the measured vibration data from an experimental rig available on the Internet.

This paper proposes a methodological approach that can be applied in practice for evaluating stakeholder dynamics and assessing projects against appropriate value propositions within an industrial maintenance project context. A conceptual framework is proposed and is demonstrated through a case analysis. It is expected that the proposed methodology, the Stakeholder Interdependent Performance Opportunities and Threats, (Stakeholder iPOT), can advance project management practice by offering a mechanism for analysing stakeholder expectations and responses to the opportunities and threats that different project events present. This study highlights the need for continued investigation not only within the context of industrial maintenance projects but also in other sectors to improve our understanding and ability to effectively manage stakeholder dynamics.

On top of the condition-based maintenance (CBM) practice for rotating machinery, the robust estimation of remaining useful life (RUL) for rolling-element bearings (REB) is of particular interest. The failure of a single bearing often results in secondary defects in the connected structure and catastrophic system failures. The prediction of RUL facilitates proactive maintenance planning to ensure system reliability and minimize financial loss due to unscheduled downtime. In this paper, to acquire early and reliable estimations of useful life, the RUL prediction of REBs is formulated into nonlinear degradation state estimation tackled by the combination of the envelope spectral indicator (ESI) and extended Kalman filter (EKF). By fusing the spectral energy of the bearing fault characteristic frequencies (FCFs) in the averaged envelope spectrum, the ESI is crafted to remove the interference from rotor-dynamics and reveal the bearing deterioration process. Once the fault is identified, the recursive Bayesian method based on EKF is utilized for estimating the bearing end-of-life time via the exponential state-space model. The distinctive advantage of the proposed approach lies in its ability to make an early prediction of RUL using a small number of ESI observations, offering an efficient practice for predictive health management at the early stage of bearing fault. The performance of the proposed method is validated using publicly available experimental bearing vibration data across three different operating conditions.

Industrial maintenance projects typically involve numerous partners with various aims, uncertainties, and risks. Therefore, effective risk and stakeholder management becomes crucial in ensuring project success. The bowtie methodology has gained recognition among organizations in various business sectors as an effective tool for analyzing and communicating potential risk scenarios. By visually illustrating how threats can lead to risk events and their consequences, this methodology can offer new insights. This paper presents earlier studies on risk and stakeholder management, key concepts in bowtie analysis, further proposing three commonalities of focus, influence, and decision-making that establish potential connections between stakeholder and risk management. Considering the potential connections and interactions between risk management and stakeholder management and with the aid of the bowtie methodology, it is expected that effective risk and stakeholder management can be promoted through enhanced stakeholder participation, risk identification and assessment, and enhancement of the implementation of efficient risk controls through visual representation and thorough analysis. Furthermore, it is expected that this integrated approach can raise stakeholder satisfaction, and enhance risk mitigation, ultimately impacting the performance of maintenance projects positively.

A robust and reliable condition monitoring and fault diagnosis system is crucial for an efficient operation of industries. Because of the advances in technologies over the past few decades, there is an increased interest in developing intelligent systems to perform tasks that traditionally rely on knowledge, experience and expertise of an individual. It is known that unexpected breakdowns have wide implications in production processes. Thus, it is vital to be able to know the machine condition and detect at the earliest possible stage the defects when they occur. Aiming at an industrial application, in this study, a two-step approach is proposed for the fault detection and diagnosis of rotor-related faults. The implemented algorithm is a pattern recognition supervised artificial neural network, which through information extracted from vibration signals allows one to identify the health status of the machine. In the first step, the model identifies whether the machine is healthy or faulty. This is important information for any industry to operate the machines. Once the machine condition (healthy or faulty) is known and if it is faulty, then only faulty machine parameters are used in the second step to know the specific fault. The model is initially based on existing experimental data, and then, it is further validated with mathematically generated data. The proposed two-step approach model and the trained framework are applied blindly at a different machine speed, where the dynamics of machine is expected to be different. The excellent results obtained suggest this approach as a possibility for industrial application.

Multi-sensor monitoring is prevalent in modern structural health monitoring (SHM) practice. As the number of sensors and sampling requirements increase, a monitoring sensor network can generate substantial data which are high-volume and high-dimensional, especially for large structure and machinery. In condition-based monitoring (CBM) of rotating machines, e.g., gas turbine, rotor fault diagnosis serves a significant role in the system reliability, safety, and efficiency, which helps reduce potential damage to the on-rotor structures and avoid catastrophic failures. For multi-channel vibration measurement, classical rotor diagnosis approaches typically involve data fusion techniques based on cross spectral analysis for identifying spectral correlation, e.g., cross power spectral density, and/or matrix analysis tool for dimensionality reduction, e.g., principal component analysis. The operation on vectors or matrices may limit their effectiveness for higher-order array or tensor data. To circumvent this limitation, the third order vibrational spectral tensor is generated by representing the multi-channel acceleration spectra as the three-way array. Through nonnegative Tucker decomposition (NTD), the spectral tensor is decomposed into multiple principal factors: the spectral factor, segmental factor, channel-wise factor, and the dense tensor core. The correlations across the characteristic spectral contents and the sensor channels are revealed by the factorization, which enables the diagnosis of rotor fault and facilitates fault localization by identifying the dominant channels of the characteristic spectral factor. The method is validated on an experimental rotor testbed where the faulty channel with crack, rub, or misalignment fault, is effectively localized via 4-channel vibration measurements, which presents a promising approach for multi-sensor fusion and fault diagnosis in rotating machinery health monitoring.

Anti-friction bearings (AFB) are crucial structural components conveying rotating motions in a variety of mechanical systems. To avoid unscheduled breakdowns and fatal failures, remaining useful life (RUL) prediction is of great practical significance in industrial practice for prognostics health management, e.g., optimizing maintenance plan for component replacements. Recently, the artificial intelligence (AI) advancements have provided effective data-driven models for bearing prognostics using machine learning. In this paper, using the variational auto-encoder (VAE) networks as the regression backbone, the bearing RUL is estimated using envelope spectra via measured vibrational data. First, the envelope spectra are utilized for bearing fault detection and the network input features. After the fault is detected, the VAE is used for learning the probabilistic mapping from the spectral input to the estimated RUL value, given its good probabilistic and generative properties over the classical auto-encoder (AE) in content generation and variational inference. The application of the method to the run-to-failure measured vibration data from the experimental rig available online have shown its efficacy in bearing RUL estimation.

An earlier study has proposed the 2-Steps approach for the rotor-related faults only using the measured vibration responses on the bearing housings on a rotating rig. The method has used the vibration parameters in both time and frequency domains calculated from the measured vibration responses at each bearing and the artificial neural network (ANN) based machine learning (ML) models to identify the machine health condition. Step-1 determines whether the machine is healthy or faulty. Then, in Step-2, only faulty conditions data are used to identify the exact nature of the faults. In the current study, this 2-Steps Approach is further extended and examined to both rotor and anti-friction bearing faults. The vibration parameters are revised to include bearing-related faults. This exercise has been done on the measured vibration data from a laboratory-scaled rotating rig. The extended study presents new insights and findings built on the previous work. This study can be a valuable contribution to predictive maintenance. It demonstrates its effectiveness in diagnosing faults in rotating machines, reducing the risk of failure, and enhancing reliability in industrial operations.