Joint Maintenance Decision Based on Remaining Useful Lifetime Prediction Using Accelerated Degradation Data

With the continuous advancement of Industry 4.0 and intelligent manufacturing, remaining useful lifetime (RUL) prediction can forecast the future degradation state of machinery and then estimate the remaining service time before it loses its safe operation ability. Accordingly, a series of predictive maintenance strategies can be regulated in advance for equipment in the Industrial Internet of Things. To tackle the challenges of insufficiency of failure data and lack of confidence in RUL prediction results, a similarity-based relevance vector machine (SRVM) is proposed in this article. Primarily, the relationship among latent variables in the SRVM is learned adaptively through similarity computations to fully utilize the limited degradation data. Furthermore, these internal variables in the SRVM are treated as time-varying variables and re-estimated dynamically to provide RUL prediction with reliable confidence. The experiment results show that the prediction accuracy of the SRVM is higher than that of other baseline methods.

This study investigates the effectiveness of embedding fiber Bragg grating (FBG) sensors in power transmission towers to assess the remaining service life of the structures following impacts from strong winds and hurricanes. FBG sensors monitor the structural integrity of the tower using online measurement of strain variations at critical structural points. The novelty of this work lies in employing a compact long short-term memory (LSTM) framework to estimate the remaining useful lifetime (RUL) from real-time FBG sensor data under both stable and fluctuating wind conditions. To estimate RUL of the tower, LSTM neural network has been implemented, providing predictive insights for proactive maintenance and risk mitigation. A prototype transmission tower was built and experimentally evaluated in a wind tunnel to assess the effectiveness and performance of the proposed RUL model. To simulate different hurricane categories, the experiment was conducted across wind speeds between 0 and 150 mph. FBG sensors installed at critical locations continuously captured real-time strain data, which was transmitted via a low-power micro FBG interrogator to a computer for input into the RUL prediction model. The proposed three-layer LSTM converges rapidly, reducing training and validation loss by nearly two orders of magnitude within 40 epochs, and achieves robust RUL predictions with an average bias of about 50 s on the test set. To quantify structural health, a mathematical health indicator was formulated based on the observed strain responses. The FBG sensors demonstrated high effectiveness in accurately detecting strain variations and monitoring the tower's dynamic behavior under extreme wind loads. The findings support the implementation of condition-based maintenance strategies, enhance safety assessments, and enable early failure detection. This approach not only improves operational reliability but also facilitates timely intervention and maintenance during critical events. Received: 26 June 2025 | Revised: 12 September 2025 | Accepted: 11 November 2025 Conflicts of Interest The authors declare that they have no conflicts of interest to this work. Data Availability Statement Data are available from the corresponding author upon reasonable request. Author Contribution Statement Yu Shi: Methodology, software, formal analysis, data curation, writing – original draft, writing – review & editing, visualization, supervision, and project administration. Abolghassem Zabihollah: Conceptualization, methodology, software, validation, investigation, resources, data curation, writing – original draft, writing – review & editing, visualization, supervision, and project administration. Yao-Chi Yu: Methodology, software, formal analysis, data curation, and writing – original draft. Arunima Pathak: Investigation, writing – original draft, and visualization. Oluwaseyi Oyetunji: Investigation.

Due to long time to get failure data in a short-time testing or even an accelerated testing conditions, light emitting diodes (LEDs) are suffering from reliability prediction problems. In the past few years, data driven approaches based on degradation data have been found interesting for degradation modeling and remaining useful lifetime (RUL) prediction of highly reliable products such as LEDs. However, many of these approaches have been implemented based on deterministic regression approaches that couldn’t handle diffusion properties and increase prediction uncertainties. In this paper, a Wiener process based method is used to model the lumen degradation and estimate the RULs of LEDs. Based on the maximum likelihood estimation, a model used to predict RUL from a degradation data has been developed. The effectiveness of proposed model for lifetime prediction is validated by using the simulation and demonstrated with the lumen flux degradation data for LEDs. The result shows that the Wiener process based method can help to estimate lifetime with better accuracy when compared to the IES-TM-21 standard with the nonlinear regression approach.

The degradation of data and failure data is used to make the maintenance decision of equipment in this paper. Firstly, the Wiener process is used to build the degradation model which is according to the degradation data of the equipment. Then, the random distribution coefficient of failure threshold is estimated by the failure data of the equipment, and we also derive the analytical expression of the remaining useful lifetime (RUL) distribution of the equipment. Finally, the maintenance decision model is established according to the renewal-reward theory and RUL prediction data which can achieve optimal maintenance strategy of equipment. The simulation example shows that the method in this paper can prolong the run time cost of equipment and effectively reduce the life cycle which has broad application prospects.

For diagnosis and predictive maintenance of mechatronic systems, monitoring of bearings is essential. An important building block for this is the determination of the bearing friction condition. This paper deals with the possibility of monitoring different journal bearing friction states, such as mixed and fluid friction, and examines a new approach to distinguish between different friction intensities under several speed and load combinations based on feature extraction and feature selection methods applied on acoustic emission (AE) signals. The aim of this work is to identify separation effective features of AE signals to subsequently classify the journal bearing friction states. Furthermore, the acquired features give information about the mixed friction intensity, which is significant for remaining useful lifetime (RUL) prediction. Time domain features as well as features in the frequency domain have been investigated in this work. To increase the sensitivity of the extracted features the AE signals were transformed to the frequency-time-domain using continuous wavelet transform (CWT). Significant frequency bands are determined to separate different friction states more effective. A support vector machine (SVM) is used to classify the signals into three different friction classes. In the end the idea for an RUL prediction method by using the already determined information is given and explained.

The demand for more reliable power conversion is ever increasing. This necessitates smart gate drivers or smart system controllers that monitor the components that are susceptible to failure. Together with the electrolytic capacitors, power semiconductor devices are among the weakest components in a power converter. In an effort to predict power MOSFET aging, this paper proposes an remaining useful lifetime (RUL) estimation algorithm for degraded power MOSFETs, which are exposed to high amplitude thermal cycles. The relative change in on-state resistance is identified as the fault signature. A data-driven RUL estimation algorithm based on a linear model approximation is proposed. The outliers present particularly in the beginning part of the data decrease the accuracy of the estimation with classical least-squares method. In this paper, a Bayesian Interference estimator is proposed to improve the accuracy through incorporating prior knowledge to estimation. The accuracy of the proposed RUL estimation tool is verified on the collected experimental data of thermally aged discrete power MOSFETs.

Accelerated degradation testing (ADT) is an efficient tool to conduct material service reliability and safety evaluations by analyzing performance degradation data. Traditional stochastic process models are mainly for linear or linearization degradation paths. However, those methods are not applicable for the situations where the degradation processes cannot be linearized. Hence, in this paper, a general ADT model based on the Wiener process is proposed to solve the problem for accelerated degradation data analysis. The general model can consider the unit-to-unit variation and temporal variation of the degradation process, and is suitable for both linear and nonlinear ADT analyses with single or multiple acceleration variables. The statistical inference is given to estimate the unknown parameters in both constant stress and step stress ADT. The simulation example and two real applications demonstrate that the proposed method can yield reliable lifetime evaluation results compared with the existing linear and time-scale transformation Wiener processes in both linear and nonlinear ADT analyses.

In this paper, we propose a practical method to estimate the milling head reliability based on the accelerated degradation test and Bayesian lifetime analysis method. By carefully analyzing the failure mechanism of the milling head, we adopt the accelerated degradation test to identify and parameterize the failure law of this component. The equivalent lifetime for the Bayesian lifetime analysis is also derived from the accelerated degradation data. Finally, the estimation is carried out based on the parametric failure law and the equivalent lifetime using the Bayesian Weibull lifetime analysis method. The reliability estimation of the milling head can be improved by the combination of the failure mechanical based accelerated degradation test and the Bayesian field lifetime data analysis. The proposed method is a practical and cost-effective reliability estimation method for highly reliable and expensive mechanical component with a pre-defined failure mechanism.

This paper discusses educational tools, resources and experience in teaching physics-of-failure (PoF) modeling and data analysis methods used in accelerated life testing (ALT) and accelerated degradation testing (ADT). The concepts introduced in this paper include several probabilistic and highly technical approaches such as the PoF and mechanistic approaches to reliability prediction and assessments as well as advanced Bayesian analysis methods used in ALT and ADT to analyze life data obtained through various methods of constant and variable stress testing efforts. Other resources include codes and algorithms from the readily available computational tools R, MATLAB, and WinBUGS for solving advanced model and life estimation methods related to PoF, ALT and ADT analyses. This paper provides an overview of these resources and several example problems.

For the highly reliable products, an accelerated degradation test (ADT) is a useful tool which has been employed in industry to obtain reliability-related information within an affordable amount of time and cost. In an ADT, as all other reliability tests, it is important to carefully design the ADT beforehand to obtain estimates of the quantities of interest as precisely as possible. In this paper, optimal ADTs are developed assuming that the constant-stress loading method is employed and the degradation characteristic follows a Wiener process. Under the constraint that the total cost does not exceed a pre-specified budget, the stress levels, the number of test units allocated to each stress level and the number of measurement (termination time) are determined such that the asymptotic variance of the maximum likelihood estimator of the q-th quantile of the lifetime distribution at the use condition is minimized.

Accelerated degradation test (ADT) has become an efficient approach to evaluate the reliability for highly reliable products. However, when modeling accelerated degradation data with degradation models, it is difficult to exactly figure out the changing rules of parameters with stress variables varying. At present, the changing rules are mainly assumed according to engineering experience or subjective judgements, which probably results in inaccurately extrapolating the reliability. To figure out the changing rules of parameters with stress variables varying, the acceleration factor constant principle and its application under ADT are studied in the paper. It is well known that the acceleration factor between any two different stress levels should be a constant under an effective ADT. For each degradation model, its parameters should obey special changing rules to satisfy that the acceleration factor maintains a constant throughout an ADT. Taking three extensively used stochastic process models as examples, including Wiener process model, gamma process model, and inverse Gaussian process model, the method of deducing the changing rules of the parameters according to the acceleration factor constant principle was demonstrated. A simulation test was conducted to validate the deduced changing rules of the parameters for the three stochastic process models. An illustrative example involving self‐regulating heating cables was used to illustrate the application of the acceleration factor constant principle under ADT. Results indicate that the acceleration factor constant principle offers an appealing and credible approach to help model accelerated degradation data. Copyright © 2016 John Wiley & Sons, Ltd.

Collecting accelerated degradation test (ADT) data can provide useful lifetime information for highly reliable products if there exists a product quality characteristic whose degradation over time can be related to reliability. However, conducting a constant-stress ADT is very costly. This is obviously not applicable for assessing the lifetime of a newly developed product because typically only a few test samples (prototypes) are available. To overcome this difficulty, we propose a step-stress accelerated degradation test (SSADT) in this paper. A case study of light emitting diodes data is used to illustrate the proposed procedure. It is seen that SSADT has several advantages over a constant-stress ADT. SSADT not only reduces the experimental cost significantly, but also provides the reliability analysts an efficient tool to assess the lifetime distribution of highly reliable products.

In reliability analysis, engineers lacks confidence to verify the mean time to failure of high power light emitting diodes from aging or degradation tests. The lifetime information, however, is important to manufacturers to improve the quality of light emitting diodes. The paper provides a bootstrap computation procedure to infer the mean time to failure of high power light emitting diodes. Assume that the degradation paths collected from an accelerated degradation test follow a Wiener process, confidence interval of the mean time to failure of high power light emitting diodes is found based on bootstrap percentiles. An illustrating example of GaN-based LEDs is used to demonstrate the use of the proposed method.

Lithium-metal batteries (LMBs) are promising candidates for accelerating the implementation of electric vehicles in our society. LMBs comprise of a lithium-metal anode that possesses the highest theoretical specific capacity (3860 mAh/g) and the lowest redox potential (-3.04V vs standard hydrogen electrode).1 LMBs undergo repeated plating and stripping of lithium ions on the anode surface. The non-homogenous stripping process leads to the isolation of metallic lithium from the anode surface, forming dead-lithium.2 The parasitic reaction between the lithium-metal anode and the electrolyte leads to the formation of the solid electrolyte interphase (SEI) layer. These phenomena lead to the reduction of available lithium and a rise in cell impedance, causing a rapid reduction of lifetime for these batteries.3 , 4 In this talk, we present a diverse dataset collected from different LMBs of different capacities and voltage ranges, that includes features from the initial formation cycle, rest voltages and CV charging during cycling. A data-driven model from the LMB cycling data is presented to predict the remaining useful lifetime (RUL) of these cells. Insights from the model can be used to determine the bad cells in a batch, improve the designs of the cells and derive optimal charging protocols for their implementation in the BMS. References Lin, D., Liu, Y. & Cui, Y. Reviving the lithium metal anode for high-energy batteries. Nat. Nanotechnol. 12, 194 (2017).Xu, S., Chen, K.-H., Dasgupta, N. P., Siegel, J. B. & Stefanopoulou, A. G. Evolution of Dead Lithium Growth in Lithium Metal Batteries: Experimentally Validated Model of the Apparent Capacity Loss. J. Electrochem. Soc. 166, A3456–A3463 (2019).Liu, G. & Lu, W. A Model of Concurrent Lithium Dendrite Growth, SEI Growth, SEI Penetration and Regrowth. J. Electrochem. Soc. 164, A1826–A1833 (2017).Gao, N. et al. Fast Diagnosis of Failure Mechanisms and Lifetime Prediction of Li Metal Batteries. Small Methods 5, 1–11 (2021).

For the accelerated degradation test (ADT) of lithium-ion battery, the degradation path models are usually applied with the fixed parameters. To optimize the degradation path models, so many battery cells are required for the ADT models based on the degradation path. However, for some lithium-ion batteries with long cycle life and high reliability, the samples are limited due to experimental cost constraints. Meanwhile, owing to the differences in raw materials and manufacturing processes, it is easy to observe variations among the degradation path of lithium-ion battery cells. Fortunately, Wiener process model considering the individual difference can express the variations and be modeled with limited samples for lithium-ion battery. Motivated by this issue, an improved method with random effects Wiener process (REWP) is proposed in this article. First, to verify the degradation process following the Wiener process, the Shapiro-Wilk test is carried out on the basis of a significance level of 0.05. Second, we analyze the varieties of every lithium-ion battery cell. To select a suitable Wiener process model, we apply the coefficient of variation as a measure of random parameters selection for Wiener process. Meanwhile, the maximum likelihood estimation (MLE) is used to optimize parameters of the proposed model. Next, the three different REWP models and general Wiener process model are compared. Finally, to demonstrate the proposed method, the lithium-ion battery cells used for a spacecraft in low Earth orbit (LEO) are conducted for ADTs. The results suggest that the REWP model has a better performance for assessing reliability of lithium-ion battery with long cycle life.