Maintenance Decision Research Articles

Framework for Identification of Bridges in Need of Monitoring Based on Maintenance History

Recently, advanced Internet-of-Things (IoT) technology has been implemented in structural health monitoring to help increase safety of infrastructure and the reliability of maintenance decisions. In this context, preliminary studies are required to identify infrastructure in need of monitoring. This study establishes a basis for the identification of bridges that require monitoring and selects the bridges for measurement in order to build an efficient IoT-based monitoring system for small and medium bridges. The bridges selected based on various maintenance histories and risk factors are expected to be utilized in comprehensively assessing the characteristics of the overall bridge network.

Key pavement condition indicators for assessing highway pavement maintenance quality index

The technical condition of road pavements requires regular monitoring to support accurate maintenance decisions. However, existing pavement maintenance quality index (PQI) assessment methods are often expensive and infrequent, which limits intelligent decision-making. This paper proposes a cost-effective and precise PQI assessment method using data from five highways in Beijing. Key PQI indicators were identified through a hybrid feature selection method, and a CatBoost model optimised by a genetic algorithm (GA-CatBoost) was developed. The model demonstrated superior accuracy, achieving an R2 of 0.938, an MSE of 0.838, and an MAE of 0.576. It achieved 98.46% accuracy compared to traditional methods, confirming its reliability in an on-site application. This approach offers an economical solution for high-frequency PQI assessment, enabling intelligent decision-making in road infrastructure maintenance and supporting the use of lightweight detection equipment.

Research on Road Internal Disease Identification Algorithm Based on Attention Fusion Mechanisms.

Internal disease in asphalt pavement is a crucial indicator of pavement health and serves as a vital basis for maintenance and rehabilitation decisions. It is closely related to the optimization and allocation of funds by highway maintenance management departments. Accurate and rapid identification of internal pavement diseases is essential for improving overall pavement quality. This study aimed to identify internal pavement diseases using deep learning algorithms, thereby improving the efficiency of determining internal pavement diseases. In this work, a multi-view recognition algorithm model based on deep learning is proposed, with attention fusion mechanisms embedded both between channels and between views. By comparing and analyzing the training and recognition results of different neural networks, it was found that the multi-view recognition algorithm model based on attention fusion demonstrates the best performance in identifying internal pavement diseases.

Remaining useful life prediction method for jointless track circuits based on multivariate feature fusion and nonlinear Wiener process

Abstract Predicting the Remaining Useful Life (RUL) of track circuits is essential to ensure the safe and reliable operation of high-speed railways. In response to the challenges faced by current machine-learning-based RUL prediction methods, which struggle to represent the uncertainty in the probability distribution of RUL predictions, this paper suggests a hybrid-driven method for estimating remaining life. Firstly, the track circuit Health Index (HI) is constructed by feature dimensionality reduction and fusion of the original multivariate monitoring data through Kernel Principal Component Analysis (KPCA) and Autoencoder (AE); Secondly, the degraded state of the rail circuit is modelled using a nonlinear Wiener degradation model. Finally, the principle of First Hitting Time (FHT) is used to derive the Probability Density Function (PDF) of the anticipated RUL. The efficacy and superiority of the approach presented in this paper are validated by experimental research on the track circuit monitoring dataset. The method enhances forecast accuracy and reduces prediction uncertainty, offering robust technical support for track circuit maintenance decision-making.

Advancing sustainable mobility: Dynamic predictive modeling of charging cycles in electric vehicles using machine learning techniques and predictive application development

The main goal in this research is to train various machine learning models to predict charging cycles in EV Electric Vehicles) battery systems. The considered models are gradient boosting, random forests, decision trees, and linear regression. Each of these was assessed based on its R-squared score, which is an important statistical measure in indicating the variance proportion yielded by the model. In contrast, the Random Forest model significantly improved, with an R-squared value of 0.83, thereby doing an excellent job in capturing nuances of the data. Only surpassed by the Gradient Boosting model at an astonishing R-squared score of 0.87, it is this excellent score that underlines its capability to predict the outcome quite accurately by modeling complex interrelations. In other words, gradient boosting outran the rest and provided the most robust results concerning drivers of students' performance. It also underlines how important choosing a good model is in educational analytics in order to increase the accuracy of the predictions. The use of these models in the proposed EV Battery Charging Cycle Predictor App results in accurate predictions to aid schedule maintenance and energy-related decisions. This research brings light to the future of advanced machine learning methods in enhancing the battery efficiencies of EVs and the development of electric mobility technologies. It is possible that the future work will imply the additional inclusion of real data and the integration of the application to general energy systems.

A benchmark on uncertainty quantification for deep learning prognostics

Reliable uncertainty quantification on RUL prediction is crucial for informative decision-making in predictive maintenance. In this context, we assess some of the latest developments in the field of uncertainty quantification for deep learning prognostics. This includes the state-of-the-art variational inference algorithms for Bayesian neural networks (BNN) as well as popular alternatives such as Monte Carlo Dropout (MCD), deep ensembles (DE), and heteroscedastic neural networks (HNN). All the inference techniques share the same inception architecture as functional model. The performance of the methods is evaluated on a subset of the large NASA N-CMAPSS dataset for aircraft engines. The assessment includes RUL prediction accuracy, the quality of predictive uncertainty, and the possibility of breaking down the total predictive uncertainty into its aleatoric and epistemic parts. Although all methods are close in terms of accuracy, we find differences in the way they estimate uncertainty. Thus, DE and MCD generally provide more conservative predictive uncertainty than BNN. Surprisingly, HNN achieve strong results without the added complexity of BNN. None of these methods exhibited strong robustness to out-of-distribution cases, with BNN and HNN methods particularly susceptible to low accuracy and overconfidence. BNN techniques presented anomalous miscalibration issues at the later stages of the system lifetime.

Resilience-Based Restoration Model for Optimizing Corrosion Repair Strategies in Tunnel Lining

In tunnel engineering, the corrosion of steel rebar is a critical factor leading to structural degradation and failure, causing a decline in load-bearing capacity, deformation, and cracking. For decision-makers, identifying the optimal timing for tunnel maintenance and selecting effective repair strategies is of paramount importance. This study introduces a resilience-based restoration model to analyze tunnel failure due to corrosion throughout its service life and to optimize the timing and selection of maintenance strategies. The model generates time-variant failure curves by constructing limit equilibrium equations. The entropy weight method is employed to quantify and weight the impact of various failure modes, determining the timing for maintenance when the failure curve exceeds a predefined threshold. Additionally, the model's uncertainty is effectively reduced through regular inspections and Bayesian updating methods, enhancing prediction accuracy. The study further incorporates a resilience index and a benefit index to provide a quantitative assessment of maintenance plans, assisting decision-makers in selecting the optimal strategy. By exemplifying the model with a case study of steel rebar corrosion in a tunnel, this paper demonstrates the model's applicability and offers a new scientific approach for quantitative maintenance decision-making in tunnel engineering.

A novel performance prediction method for harmonic reducers based on the trend-enhanced Fisher’s discriminant ratio-Wiener process

Abstract Harmonic reducers, as core components of industrial robots, play a critical role in maintaining robot health. Performance degradation assessment (PDA) and remaining useful life (RUL) prediction are essential for ensuring the operational reliability of such robots. To address the multistage characteristics and stochastic uncertainties in the performance degradation of harmonic reducers, a performance prediction method based on Fisher’s discriminant ratio and Wiener process is proposed. Firstly, the health indicator (HI) construction is defined as an optimization problem based on Fisher’s discriminant ratio and trend monotonicity constraints. By leveraging the sum of the multiplication of frequency amplitudes and optimized weights, precise segmentation of the multistage degradation process over the entire lifecycle is achieved. Notably, the optimized weights can automatically identify the resonance frequency bands caused by damage. Subsequently, a nonlinear Wiener process model with a drift coefficient in the form of a power function is established. The probability density function (PDF) expression for RUL is derived based on the concept of first hit time (FHT). Additionally, a logarithmic likelihood function (Log-LF) for the unknown parameters in the degradation model is constructed. The experimental results indicate that the proposed method surpasses the other two Wiener process models in terms of root mean square error (RMSE), mean absolute percentage error (MAPE), and cumulative relative accuracy (CRA). This provides robust support for preventive maintenance decision-making for harmonic reducers.

Digital twin-enhanced opportunistic maintenance of smart microgrids based on the risk importance measure

Smart microgrids face more diverse and frequent risks than traditional grids due to their complexity and reliance on distributed generation. Ensuring the reliable operation of smart microgrids requires effective maintenance. Traditional maintenance methods, based on periodic inspections and fault diagnosis, struggle to adapt to the dynamics and complexity of microgrid systems. The introduction of digital twin technology provides a new solution for the opportunistic maintenance of microgrid systems. This paper presents a digital twin microgrid architecture for real-time monitoring and decision-making in opportunistic maintenance. Meanwhile, this paper introduces a risk importance measure to aid to optimize opportunistic maintenance strategies when resources are limited. Finally, a wind-solar-storage microgrid is used to illustrate the proposed method. Experimental results show that the proposed method significantly reduces maintenance costs and improves system reliability, effectively supporting microgrid maintenance.

State-of-health estimation for lithium-ion batteries using relaxation voltage under dynamic conditions

The data-driven approach accurately estimates the state-of-health of lithium-ion batteries using online data, aiding consumers in operational and maintenance decisions. However, the stochastic charging and discharging behavior in realistic scenarios leads to continuous transient processes that render conventional features undetectable or exacerbate fluctuations. Here, we use domain knowledge and equivalent circuit modeling to investigate the extraction of physical features of aging through a relatively stable relaxation process under dynamic conditions. Our study uses 16-cell data from the National Aeronautics and Space Administration randomized dataset and compares four basic data-driven models for validation. The results show that incorporating a limited set of previous discharge step information significantly enhances model robustness and accuracy. The best-performing model, auto-relevance determination gaussian process regression, achieves a low root mean square error of 1.94 %. Physically interpretable features do not rely on historical data, require a smaller sample size, and exhibit greater generalizability across different current scenarios. This method does not depend on a specific charging method, making it practical and adaptable. Therefore, the data-driven approach utilizing relaxation voltages and correlation features offers a viable solution for accurately estimating the health state of lithium-ion batteries under dynamic conditions.

Data augmentation using SMOTE technique: Application for prediction of burst pressure of hydrocarbons pipeline using supervised machine learning models

Accurate burst pressure prediction is critical for ensuring oil and gas pipeline safety, guiding maintenance decisions, and lowering costs and risks. Traditional methods have limitations, including high experimental costs, conservative empirical models, and computationally expensive numerical algorithms. Machine learning (ML) models have supplanted traditional methods in recent years. However, small and imbalanced datasets are the big challenge to build a ML model that can generate more accurate results. Moreover, the lack of generalization in ML models trained on a dataset of pipelines with specific material grids prevents them from producing superior results on other pipeline types. First, FEA was used to make a dataset. Then, a new way to improve machine learning (ML) model generalization for burst pressure prediction is suggested: combine publicly available datasets of different pipeline specifications. In this combined dataset, some pipelines have a higher number of data samples, and some have fewer, which causes a class imbalance issue. The Synthetic Minority Oversampling Technique (SMOTE) technique was applied to address the issue of class imbalance. The performance of various ML models, Extra Trees (ET), Extreme Gradient Boosting (XGBR), Random Forest (RF), Light Gradient Boosting Machine (LGBM), and Decision Tree (DT), was evaluated to validate the model's prediction and generalization on pipelines of various material grids. Results show that all the selected ML models produced high R-squared, i.e., >0.95, on balanced data compared to the imbalance dataset. These results show that SMOTE-based augmentation is a beneficial way to fix dataset imbalance and make ML models better at predicting burst pressure in oil and gas pipelines.

Application of vibration signals in railway track diagnostics using a mobile railway platform

The article presents a comprehensive method for using vibration signals to diagnose railway tracks. The primary objective is to gather detailed information on track conditions through a passive experiment. This involves using mobile diagnostic tools and techniques to assess railway infrastructure. The article elaborates on the range of diagnostic activities conducted in accordance with detailed railway regulations and highlights the benefits and capabilities of mobile diagnostics in railway transport. The research includes mobile field measurements across the general railway manager’s network, employing vibration signals to detect and evaluate track conditions. The methodology section provides a thorough description of the mobile measurement rail platform, detailing the equipment used, the routes taken for measurements, and the processes of data acquisition and processing. The data obtained from these measurements is crucial for understanding the actual technical condition of the railway tracks. The method of obtaining and processing data is explained in relation to the real technical condition of the railway track. This involves using transducers with specific parameters and parametrically defined signal recording, along with dedicated analysis techniques in post-processing. Vibration signals serve as the primary carrier of information in this diagnostic method. The article details the step-by-step procedures for collecting and analyzing these signals to provide accurate assessments of track conditions. Based on the results from the mobile measurement rail platform, the article characterizes various areas of diagnostics where vibration signals are particularly effective for technical evaluation. These areas include identifying track defects, monitoring track surface and railway crossing and assessing the overall structural health of the railway infrastructure. The use of vibration signals offers a non-invasive and efficient means of track diagnostics, providing real-time data for maintenance and repair decisions. In conclusion, the article underscores the significance of mobile diagnostics in enhancing the safety and reliability of railway transport. By leveraging vibration signals and advanced data processing techniques, this method provides a framework for continuous monitoring and assessment of railway track conditions, ultimately contributing to improved maintenance strategies and operational efficiency.

Acoustic emission failure detection and localisation in historic fibrous plaster ceiling

Fibrous plaster (FP) is a composite material that has been used for suspended decorative ceilings in historical buildings since the late 19th century. Such ceilings are susceptible to failures, particularly in hard-to-access structural hangers (‘wads’). Due to a limited understanding of structural behaviour of FP ceilings, maintenance and reconstruction decisions are based on visual and tactile inspections. Such inspections are costly, intrusive and ultimately subjective. Non-destructive techniques are needed to detect and localise failures in FP ceilings. An Acoustic Emission system was recently developed to detect and localise damage sources in FP ceilings. The method processes data from AE sensor clusters with one-dimensional convolutional neural networks(1D-CNNs) trained to account for wave propagation anisotropy. This paper presents a case study conducted at the Institution of Civil Engineers (ICE) building in London (UK) to validate the method. The results confirm the ability of the method to localise failures under real site conditions in 450 mm-by-450 mm monitoring areas. The method may enable non-intrusive detection of failures from the underside of ceilings during periodic inspections. The possibility of conducting pre-training in laboratories and covering large monitoring areas up to 1500mm-by-1500mm is established and may facilitate practical applications in the future.

Enhanced vertical railway track quality index with dynamic responses from moving trains

The conventional vertical track quality index (TQI) based on the standard deviation of longitudinal levels yields standardized railway track condition assessment. Nevertheless, its capability to identify problems is limited, particularly in the ballast and substructure layers when abrupt changes affect train-track interaction. Previous research shows that dynamic responses from moving trains via axle box acceleration (ABA) measurements can quantify abrupt changes in the vertical dynamic responses. Thus, this paper proposes a framework to design an enhanced vertical TQI, called EnVTQI, by integrating track longitudinal levels and dynamic responses from ABA measurements. First, measured ABA signals are processed to mitigate the influence of variation in measurement speed. Then, substructure and ballast-related features are extracted, including scale average wavelet power (SAWP) in the ranges 0.04 m-1 to 0.33 m-1 (substructure) and 1.25 m-1 to 2.50 m-1 (ballast). This enables identifying track conditions at different track layers. Finally, EnVTQI is determined by weight averaging between the conventional vertical TQI and the ABA features from moving trains. The performance of EnVTQI is evaluated based on 48 segments of a 200-m track on a Dutch railway line. The results indicate that EnVTQI helps to distinguish track segments that cause poor train-track interaction, which the conventional TQI does not indicate. EnVTQI can supplement the conventional TQI, improving the effectiveness of track maintenance decision-making.

Fatigue research of metro car body based on operational loads

PurposeThe principle of infinite life design currently directs fatigue resistance strategies for metro car bodies. However, this principle might not fully account for the dynamic influence of operational loads and the inevitable presence of defects. This study aims to integrate methods of service life estimation and residual life assessment, which are based on operational loads, into the existing infinite life verification framework to further ensure the operational safety of subway trains.Design/methodology/approachOperational loads and fatigue loading spectra were determined through the field test. The material test was conducted to investigate characteristics of the fracture toughness and the crack growth rate. The fatigue strength of the metro car body was first verified using the finite element method and Moore–Kommers–Japer diagrams. The service life was then estimated by applying the Miner rule and high-cycle fatigue curves in a modified form of the Basquin equation. Finally, the residual life was assessed utilizing a fracture assessment diagram and a fitted curve of crack growth rate adhered to the Paris formula.FindingsNeither the maximum utilization factor nor the cumulative damage exceeds the threshold value of 1.0, the metro car body could meet the design life requirement of 30 years or 6.6 million km. However, three out of five fatigue key points were significantly influenced by the operational loads, which indicates that a single fatigue strength verification cannot achieve the infinite life design objective of the metro car body. For a projected design life of 30 years, the tolerance depth is 12.2 mm, which can underscore a relatively robust damage tolerance capability.Originality/valueThe influence of operational loads on fatigue life was presented by the discrepancy analysis between fatigue strength verification results and service life estimation results. The fracture properties of butt-welded joints were tested and used for the damage tolerance assessment. The damage tolerance life can be effectively related by a newly developed equation in this study. It can be a valuable tool to provide the theoretical guidance and technical support for the structural improvements and maintenance decisions of the metro car body.

Multi-Criteria Decision-Making Framework (AHP-TOPSIS): Pavement Preventive Maintenance Case Study for Ordinary National Trunk Highways

Pavement maintenance and rehabilitation decision-making needs to weigh multiple strategic goals to achieve sustainable development through the pavement maintenance management system. Making decisions regarding pavement preventive maintenance is both intricate and costly. This study introduces a multi-criteria decision-making framework aimed at enhancing the scientific basis of such decisions. The framework first establishes an evaluation system for preventive maintenance strategies by considering three primary evaluation criteria—service functionality, pavement performance, and economic benefits, and then identifies nine specific evaluation indicators to influence these criteria, with a comparison matrix constructed to determine the weight of each indicator in relation to the maintenance decision hierarchy. Following this, the technique for order preference by similarity to ideal solution (TOPSIS) is employed to prioritize four commonly utilized preventive maintenance strategies. The results reveal that pavement condition and maintenance costs are the most influential factors in determining the appropriate preventive maintenance strategies for national highways. The priority rankings for the four strategies—slurry seal, micro-surfacing, chip seal, and ultra-thin overlays—are found to be 56.12%, 63.86%, 12.12%, and 83.52%, respectively, with ultra-thin overlays identified as the optimal choice for second-class highways. The decision-making model utilized in this study enables a multi-dimensional analysis, reducing the subjectivity inherent in expert evaluations and facilitating the prompt identification of the most suitable maintenance strategy.

A novel dynamic predictive maintenance framework for gearboxes utilizing nonlinear Wiener process

Abstract In the context of advancing industrial automation, gearboxes, as pivotal components in power transmission systems, have a direct bearing on the operational efficiency and safety of the entire machinery. This study introduces a novel dynamic predictive maintenance framework for gearboxes using a nonlinear Wiener process. Comprehensive experiments validate the framework, demonstrating significant reductions in maintenance costs and improvements in reliability. First, a full-life degradation experiment was executed on the gearbox, leveraging the root mean square (RMS) value of the vibration signal as an indicator of system degradation. Subsequently, the signals from four vibration sensors were synthesized and normalized through Kernel Principal Component Analysis (KPCA), thereby enabling a more nuanced representation of the gearbox's degradation profile. The degradation trajectory was then modeled using a nonlinear Wiener process framework. The Wiener process's parameters and state variables were iteratively refined utilizing an online filtering algorithm grounded in Bayesian inference. This facilitated the derivation of the probability density function for the remaining useful life (RUL), thereby enabling a robust prediction of the gearbox's RUL. Finally, to minimize maintenance costs per unit of time, an optimization model for dynamic maintenance decision-making was formulated. The optimal maintenance timing was ascertained by solving this model. The empirical findings of this investigation demonstrate the efficacy of the proposed approach in executing dynamic predictive maintenance for gearboxes. This research endeavors to furnish novel theoretical underpinnings and pragmatic directives for the field of predictive maintenance in the context of gearboxes.

Multi-objective predictive maintenance scheduling models integrating remaining useful life prediction and maintenance decisions

This paper presents two novel data-driven multi-objective predictive maintenance scheduling models that integrate remaining useful life (RUL) prediction into maintenance planning. First, we propose a deep learning ensemble model that includes a convolutional neural network and bidirectional long short-term memory network with a temporal self-attentional mechanism and Bayesian optimization method to predict system RUL effectively. Then, two novel multi-objective mixed-integer linear programming (MILP) models are developed based on the system-predicted RUL to deal with the system predictive maintenance problem, which aims to minimize the total maintenance completion time and maintenance-related costs simultaneously. To solve this problem, we design an iteration ϵ-constraint method to achieve Pareto solutions. Meanwhile, a fuzzy logic method is proposed to recommend a preferred Pareto solution for decision-makers. Finally, the aircraft engine C-MAPSS dataset from NASA validates that the effectiveness of the proposed data-driven multi-objective predictive maintenance scheduling models are effective. For the predictive maintenance of 20 aircraft engines, both the maintenance completion time and maintenance cost are reduced by more than 40%.

Enhancing power equipment defect identification through multi-label classification methods

Accurate identification and classification of equipment defects are essential for assessing the health of power equipment and making informed maintenance decisions. Traditional defect classification methods, which rely on subjective manual records and intricate defect descriptions, have proven to be inefficient. Existing approaches often evaluate equipment status solely based on defect grade. To enhance the precision of power equipment defect identification, we developed a multi-label classification dataset by compiling historical defect records. Furthermore, we assessed the performance of 11 established multi-label classification methods on this dataset, encompassing both traditional machine learning and deep learning methods. Experimental results reveal that methods considering label correlations exhibit significant performance advantages. By employing balanced loss functions, we effectively address the challenge of sample imbalance across various categories, thereby enhancing classification accuracy. Additionally, segmenting the power equipment defect classification task, which involves numerous labels, into label recall and ranking stages can substantially improve classification performance. The dataset we have created is available for further research by other scholars in the field.

Grouping-maintenance of multi-components production system under ecological background

Purpose: The purpose of this paper is to propose an opportunistic group maintenance model for a multi-component series system considering reducing CO2 emission and increasing system efficiency, which establishes a maintenance policy to minimize the cost rate of the system life cycle.Design/methodology/approach: Structural dependence and economic dependence between the components in a multi-component series system are analyzed to make a condition-based maintenance policy. To minimize the cost rate of the system life cycle, clustering theory and two decision variables, which include the preventative maintenance cycle multiplier and basic preventive maintenance interval of each component, is utilized to make an opportunistic policy for system optimal maintenance under the background of increasing energy efficiency and reducing CO2 emissions.Findings: It can be concluded that the government imposes fines on excessive CO2 emissions and energy consumption indicators, which can influence the maintenance decisions of enterprises, promote enterprises to shorten the cycle of preventive maintenance, and avoid excessive CO2 emissions of various components and exceed the indicators as much as possible, so as to enable enterprises to actively save energy, reduce emissions and control carbon emissions. Furthermore, when the single preparation cost of repair maintenance increases, enterprises need to shorten the maintenance period to avoid frequent component failures. As the cost of a single prep for preventive maintenance rises, organizations need to extend maintenance cycles and avoid frequent parts downtime - minimizing their own repair costs.Practical implications: Considering the high maintenance cost and low energy efficiency of multi-component systems, this model assists production managers to have better maintenance of these systems.Originality/value: 1.Model innovation: comprehensive consideration of two variables in the selection of decision variables for preventive maintenance or opportunistic maintenance; The selection and synthesis of ecological factors when making ecologically conscious maintenance decisions for multi-component systems make up for the gaps of single variables and one-sided indicators in previous studies and models. 2. Methodological innovation: In the identification of opportunity maintenance, the idea of clustering is first combined; In the simulation analysis, genetic algorithm is used to obtain the optimal parameters quickly and accurately. A blend of management, statistics, biology and computer science.