The diagnostic accuracy of analytical models for fault section diagnosis of power systems relies heavily on the correction of protective relays (PRs) and circuit breakers (CBs). The current analytical models use the received alarm information directly, but the actions of PRs and CBs are fraught with uncertainties of mal-operation and miss-operation, and they are also subject to change during the uploading process, which may result in wrong results. To address this issue, this study presents an information correction method to correct those wrong or unreasonable PRs and CBs. Different abnormal action situations of PRs and CBs for busbars, lines, and transformers are considered and used to derive the corresponding correction strategies. Besides, an improved biogeography-based optimization based on binary coding and Boolean operations is developed to solve the analytical model. Simulations on two power systems indicate the accuracy of the analytical model and the superiority of the solving method.
{"title":"Information Correction–Based Analytical Model for Fault Section Diagnosis of Power Systems","authors":"Guojiang Xiong;Shunshun Sun","doi":"10.1109/TR.2025.3549059","DOIUrl":"https://doi.org/10.1109/TR.2025.3549059","url":null,"abstract":"The diagnostic accuracy of analytical models for fault section diagnosis of power systems relies heavily on the correction of protective relays (PRs) and circuit breakers (CBs). The current analytical models use the received alarm information directly, but the actions of PRs and CBs are fraught with uncertainties of mal-operation and miss-operation, and they are also subject to change during the uploading process, which may result in wrong results. To address this issue, this study presents an information correction method to correct those wrong or unreasonable PRs and CBs. Different abnormal action situations of PRs and CBs for busbars, lines, and transformers are considered and used to derive the corresponding correction strategies. Besides, an improved biogeography-based optimization based on binary coding and Boolean operations is developed to solve the analytical model. Simulations on two power systems indicate the accuracy of the analytical model and the superiority of the solving method.","PeriodicalId":56305,"journal":{"name":"IEEE Transactions on Reliability","volume":"74 3","pages":"3847-3855"},"PeriodicalIF":5.7,"publicationDate":"2025-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144998368","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Zhiqiang Li;Hongyu Zhang;Xiao-Yuan Jing;Wangyang Yu;Yueyue Liu
The core goal of software defect prediction (SDP) is to identify modules with a high likelihood of defects, thereby enabling prioritization of quality assurance activities with low inspection effort. There are many supervised defect prediction models that are extensively studied. However, these methods require the need for labeling data to get enough training modules, which will cause a lot of waste of human resources. Cross-project defect prediction primarily reuses models trained on other projects with enough historical data. However, this strategy is often hindered by large distribution differences across different projects and privacy concerns of data. Unsupervised learning technique is an alternative solution to the unlabeled data, but it mainly focuses on single-view prediction by concatenating all the software metrics. This ignores the diversity and complementarity of different types of metrics. This study proposes a novel approach, namely, multiview unsupervised software defect prediction (MUSDP). It aims to collaboratively learn the diversity and complementarity of different views to build a robust and reliable defect prediction model. Extensive experiments on $ 28$ releases from eight software projects indicate that MUSDP exhibits superior or comparable results regarding G-mean, AUC, $P_{text{opt}}$, and Recall@20% compared to competing supervised and unsupervised methods. For the interpretation of MUSDP, the number of added and deleted lines significantly influence its predictions.
{"title":"Unsupervised Software Defect Prediction Through Multiview Clustering","authors":"Zhiqiang Li;Hongyu Zhang;Xiao-Yuan Jing;Wangyang Yu;Yueyue Liu","doi":"10.1109/TR.2025.3548107","DOIUrl":"https://doi.org/10.1109/TR.2025.3548107","url":null,"abstract":"The core goal of software defect prediction (SDP) is to identify modules with a high likelihood of defects, thereby enabling prioritization of quality assurance activities with low inspection effort. There are many supervised defect prediction models that are extensively studied. However, these methods require the need for labeling data to get enough training modules, which will cause a lot of waste of human resources. Cross-project defect prediction primarily reuses models trained on other projects with enough historical data. However, this strategy is often hindered by large distribution differences across different projects and privacy concerns of data. Unsupervised learning technique is an alternative solution to the unlabeled data, but it mainly focuses on single-view prediction by concatenating all the software metrics. This ignores the diversity and complementarity of different types of metrics. This study proposes a novel approach, namely, multiview unsupervised software defect prediction (MUSDP). It aims to collaboratively learn the diversity and complementarity of different views to build a robust and reliable defect prediction model. Extensive experiments on <inline-formula><tex-math>$ 28$</tex-math></inline-formula> releases from eight software projects indicate that MUSDP exhibits superior or comparable results regarding <italic>G-mean</i>, <italic>AUC</i>, <inline-formula><tex-math>$P_{text{opt}}$</tex-math></inline-formula>, and <italic>Recall@20%</i> compared to competing supervised and unsupervised methods. For the interpretation of MUSDP, the number of added and deleted lines significantly influence its predictions.","PeriodicalId":56305,"journal":{"name":"IEEE Transactions on Reliability","volume":"74 3","pages":"3356-3370"},"PeriodicalIF":5.7,"publicationDate":"2025-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144997922","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
In this article, a two-step approach is developed to estimate mean time to failure (MTTF) of solid-state drives (SSD) by first formulating a composite health indicator via multichannel signal fusion and further predicting the remaining useful life(RUL) under degradation model misspecification. Specifically, an unsupervised neural network based on self-organizing map is constructed to approximate the highly nonlinear relationship between multivariate monitoring attributes and a univariate SSD health indicator. For each SSD, the composite health indicator over time is further calibrated by smoothing techniques and formulated into a general path degradation model with a uniform failure threshold. By extrapolating each degradation path to hit the failure threshold, the RULs of SSDs are obtained as pseudofailure times, which are fitted by various lifetime distributions. Finally, a novel model averaging strategy is proposed to weigh the MTTFs estimated by multiple combinations of candidate degradation models and lifetime distributions to alleviate the impact of model misspecification. A real-world SSD dataset is used to demonstrate the feasibility of the proposed two-step approach. Numerical results suggest that the proposed approach better characterizes the underlying degradation process under different model assumptions and settings.
{"title":"Estimating Mean Time to Failure of Solid-State Drives via Self-Organizing Map and Model Averaging","authors":"Peng Li;Xun Xiao;Jiayu Chen","doi":"10.1109/TR.2025.3550380","DOIUrl":"https://doi.org/10.1109/TR.2025.3550380","url":null,"abstract":"In this article, a two-step approach is developed to estimate mean time to failure (MTTF) of solid-state drives (SSD) by first formulating a composite health indicator via multichannel signal fusion and further predicting the remaining useful life(RUL) under degradation model misspecification. Specifically, an unsupervised neural network based on self-organizing map is constructed to approximate the highly nonlinear relationship between multivariate monitoring attributes and a univariate SSD health indicator. For each SSD, the composite health indicator over time is further calibrated by smoothing techniques and formulated into a general path degradation model with a uniform failure threshold. By extrapolating each degradation path to hit the failure threshold, the RULs of SSDs are obtained as pseudofailure times, which are fitted by various lifetime distributions. Finally, a novel model averaging strategy is proposed to weigh the MTTFs estimated by multiple combinations of candidate degradation models and lifetime distributions to alleviate the impact of model misspecification. A real-world SSD dataset is used to demonstrate the feasibility of the proposed two-step approach. Numerical results suggest that the proposed approach better characterizes the underlying degradation process under different model assumptions and settings.","PeriodicalId":56305,"journal":{"name":"IEEE Transactions on Reliability","volume":"74 3","pages":"4417-4425"},"PeriodicalIF":5.7,"publicationDate":"2025-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144998038","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This article tries to address the concerns about remaining useful life (RUL) prediction across machines: 1) what data from source domain contributes more to transfer prediction? and 2) is the information transfer reliable enough? This article proposes a novel fault mode-oriented deep tensor domain-adversarial regression adaptation approach to achieve interpretable RUL transfer prediction across machines. First, by integrating fault mechanism and degradation characteristics, a new fault mode-oriented significance indicator (FSI) is constructed based on tensor representation to evaluate the importance of degradation data from source domain. Second, a multisubdomains adversarial regression adaptation network, in which each subsource domain corresponds to a fault mode, is constructed to purposefully transfer the degradation knowledge from source domain. The domain discriminator for each subsource domain is adaptively weighted by FSIs that are updated in each round of adversarial training. An alternating optimization algorithm is then designed to find the optimal knowledge representation and transfer effect. Moreover, an upper bound of prediction error is derived for the proposed approach, which offers a theoretical guarantee for cross-machine prognostic task. Experimental results on three benchmark datasets empirically validate the proposed approach under fixed and varying working conditions, and can reveal fault modes' significance for more trustworthy prediction.
{"title":"An Interpretable and Reliable Remaining Useful Life Prediction Approach Across Different Machines With Tensor Domain-Adversarial Regression Adaptation","authors":"Wentao Mao;Jiayi Wang;Wen Zhang;Yuan Li;Panpan Zeng;Zhidan Zhong","doi":"10.1109/TR.2025.3547426","DOIUrl":"https://doi.org/10.1109/TR.2025.3547426","url":null,"abstract":"This article tries to address the concerns about remaining useful life (RUL) prediction across machines: 1) what data from source domain contributes more to transfer prediction? and 2) is the information transfer reliable enough? This article proposes a novel fault mode-oriented deep tensor domain-adversarial regression adaptation approach to achieve interpretable RUL transfer prediction across machines. First, by integrating fault mechanism and degradation characteristics, a new fault mode-oriented significance indicator (FSI) is constructed based on tensor representation to evaluate the importance of degradation data from source domain. Second, a multisubdomains adversarial regression adaptation network, in which each subsource domain corresponds to a fault mode, is constructed to purposefully transfer the degradation knowledge from source domain. The domain discriminator for each subsource domain is adaptively weighted by FSIs that are updated in each round of adversarial training. An alternating optimization algorithm is then designed to find the optimal knowledge representation and transfer effect. Moreover, an upper bound of prediction error is derived for the proposed approach, which offers a theoretical guarantee for cross-machine prognostic task. Experimental results on three benchmark datasets empirically validate the proposed approach under fixed and varying working conditions, and can reveal fault modes' significance for more trustworthy prediction.","PeriodicalId":56305,"journal":{"name":"IEEE Transactions on Reliability","volume":"74 3","pages":"4076-4090"},"PeriodicalIF":5.7,"publicationDate":"2025-03-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144998360","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Jung-San Lee;Tzu-Hao Chen;Chit-Jie Chew;Po-Yao Wang;Yun-Yi Fan
Zero-trust architecture has received massive attention globally and been a significant development in the field of cybersecurity. Within zero-trust architecture, the continuous authentication (CA) strategy has been proposed to counter the network security threats posed by traditional static authentication mechanisms. However, most studies have focused on either device-to-device authentication or user authentication. This limitation results in risks of identity spoofing or credential theft despite the implementation of the CA mechanism, thus concluding the parity in significance between authenticating users and devices. Furthermore, considering the CA of users, it is essential to face the issue posed by user authentication fatigue. In response to these challenges, this work aims to introduce an unconsciously CA protocol (UCAP) based on zero-trust concepts and behavior biometrics. UCAP utilizes the behavior of keystroke dynamics as a main factor in consistently evaluating the user trust level. This method enables the continual updating of communication keys to preserve robust authentication of both devices and users. The robustness of UCAP has been examined through formal tools, while the experimental outcomes have shown satisfactory performance.
{"title":"Unconsciously Continuous Authentication Protocol in Zero-Trust Architecture Based on Behavioral Biometrics","authors":"Jung-San Lee;Tzu-Hao Chen;Chit-Jie Chew;Po-Yao Wang;Yun-Yi Fan","doi":"10.1109/TR.2025.3541224","DOIUrl":"https://doi.org/10.1109/TR.2025.3541224","url":null,"abstract":"Zero-trust architecture has received massive attention globally and been a significant development in the field of cybersecurity. Within zero-trust architecture, the continuous authentication (CA) strategy has been proposed to counter the network security threats posed by traditional static authentication mechanisms. However, most studies have focused on either device-to-device authentication or user authentication. This limitation results in risks of identity spoofing or credential theft despite the implementation of the CA mechanism, thus concluding the parity in significance between authenticating users and devices. Furthermore, considering the CA of users, it is essential to face the issue posed by user authentication fatigue. In response to these challenges, this work aims to introduce an unconsciously CA protocol (UCAP) based on zero-trust concepts and behavior biometrics. UCAP utilizes the behavior of keystroke dynamics as a main factor in consistently evaluating the user trust level. This method enables the continual updating of communication keys to preserve robust authentication of both devices and users. The robustness of UCAP has been examined through formal tools, while the experimental outcomes have shown satisfactory performance.","PeriodicalId":56305,"journal":{"name":"IEEE Transactions on Reliability","volume":"74 2","pages":"2591-2604"},"PeriodicalIF":5.0,"publicationDate":"2025-03-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144205967","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The problem of dc interference caused by stray currents in dc traction power supply system (DPS) is becoming increasingly serious. In order to study the interference degree of stray currents, a unified model of DPS and stray current dissipation based on the direct boundary element method (UBEM) has been established. The stray current collection network (SCCN) polarization potential is an important index to evaluate the leakage level of stray current. In this article, the relationship between SCCN polarization potential and rail-to-earth resistance (RE), train headways and longitudinal resistance of SCCN is investigated. It provides a partial theoretical basis and calculation method for stray current protection and system optimization. Field tests and CDEGS software simulations prove that UBEM is effective. The results show that UBEM is within 6.07% of the CDEGS simulation results and within 11.45% of the field test results. Taking the actual metro project in China as an example, SCCN polarization potential is only affected by local stray current. When RE>7.35 Ω·km, The average value of the SCCN polarization potential drops below 0.5 V.
{"title":"A Unified Model of DC Traction Power Supply System and Stray Current Dissipation","authors":"Wei Liu;Feilong Liu;Zhe Pan;Zhuoxin Yang;Jianbang Niu","doi":"10.1109/TR.2025.3546090","DOIUrl":"https://doi.org/10.1109/TR.2025.3546090","url":null,"abstract":"The problem of dc interference caused by stray currents in dc traction power supply system (DPS) is becoming increasingly serious. In order to study the interference degree of stray currents, a unified model of DPS and stray current dissipation based on the direct boundary element method (UBEM) has been established. The stray current collection network (SCCN) polarization potential is an important index to evaluate the leakage level of stray current. In this article, the relationship between SCCN polarization potential and rail-to-earth resistance (RE), train headways and longitudinal resistance of SCCN is investigated. It provides a partial theoretical basis and calculation method for stray current protection and system optimization. Field tests and CDEGS software simulations prove that UBEM is effective. The results show that UBEM is within 6.07% of the CDEGS simulation results and within 11.45% of the field test results. Taking the actual metro project in China as an example, SCCN polarization potential is only affected by local stray current. When RE>7.35 Ω·km, The average value of the SCCN polarization potential drops below 0.5 V.","PeriodicalId":56305,"journal":{"name":"IEEE Transactions on Reliability","volume":"74 3","pages":"4194-4206"},"PeriodicalIF":5.7,"publicationDate":"2025-03-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144998137","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
As a distinctive redundant form in various practical applications, load-sharing systems consist of stochastically dependent units bearing system load altogether. Conventional load-sharing systems usually operate under an equal load allocation policy, and the system load is evenly distributed among all working units. However, this static policy neglects the individual dynamic and heterogenous characteristics during unit degradation processes, and leads to nonnegligible individual differences between unit reliability and lifetime distributions. Faced with this problem, this article proposes a novel condition-based operation and maintenance strategy for two-unit load-sharing systems. Each unit undergoes nonmonotonic continuous degradation following the Wiener process, and the system reliability is evaluated by considering a possible two-phase degradation process of the surviving unit once one unit fails. At each periodic inspection time, the system load is dynamically allocated by minimizing the Jensen–Shannon divergence between unit remaining useful lifetime distributions. Furthermore, a condition-based maintenance model is established according to semi-renewal process characteristics, along with specific theoretical analysis for the stationary distribution of system states. Compared with traditional operation and maintenance strategies, the effectiveness of the proposed strategy is validated through numerical experiments, and a practical case study of a two-cell lithium-ion battery pack illustrates robust economic benefit in dynamically adjusting the battery cell loads.
{"title":"Condition-Based Operation and Maintenance Strategy for Load-Sharing Systems Based on Wiener Process","authors":"Wei Chen;Songhua Hao","doi":"10.1109/TR.2025.3545037","DOIUrl":"https://doi.org/10.1109/TR.2025.3545037","url":null,"abstract":"As a distinctive redundant form in various practical applications, load-sharing systems consist of stochastically dependent units bearing system load altogether. Conventional load-sharing systems usually operate under an equal load allocation policy, and the system load is evenly distributed among all working units. However, this static policy neglects the individual dynamic and heterogenous characteristics during unit degradation processes, and leads to nonnegligible individual differences between unit reliability and lifetime distributions. Faced with this problem, this article proposes a novel condition-based operation and maintenance strategy for two-unit load-sharing systems. Each unit undergoes nonmonotonic continuous degradation following the Wiener process, and the system reliability is evaluated by considering a possible two-phase degradation process of the surviving unit once one unit fails. At each periodic inspection time, the system load is dynamically allocated by minimizing the Jensen–Shannon divergence between unit remaining useful lifetime distributions. Furthermore, a condition-based maintenance model is established according to semi-renewal process characteristics, along with specific theoretical analysis for the stationary distribution of system states. Compared with traditional operation and maintenance strategies, the effectiveness of the proposed strategy is validated through numerical experiments, and a practical case study of a two-cell lithium-ion battery pack illustrates robust economic benefit in dynamically adjusting the battery cell loads.","PeriodicalId":56305,"journal":{"name":"IEEE Transactions on Reliability","volume":"74 3","pages":"4402-4416"},"PeriodicalIF":5.7,"publicationDate":"2025-03-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144998247","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Shraddha Tripathi;Faheem Nizar;Om Jee Pandey;Tushar Sandhan;Rajesh M. Hegde
The Internet of Things (IoT) has grown explosively with wireless technology integration. Several IoT applications require high data throughput, low data transmission latency, and high data gathering reliability. Since, the IoT network (IoTN) is generally dynamic and utilizes a multi-hop data transmission scheme for such applications, the throughput, latency, and network lifetime tend to degrade as the hops increase. Moreover, IoT devices (IoD) are low-cost, less computationally capable, and battery-limited, further impacting performance. A faulty IoD worsens network lifetime and throughput. Predicting faulty nodes and re-routing data can significantly enhance performance. This work proposes a node fault prediction framework to enhance data routing in dynamic IoTN, maximizing throughput and lifetime. The network is represented as a graph in which the IoD are the nodes. Then a novel deep learning model is proposed utilizing various node and edge features to predict the faulty IoDs. Particularly, the proposed edge and node features-accumulation deep learning (ENADL) method exploits features, such as Euclidean distance between nodes, residual energy level of nodes, and type and number of messages passed between edges to predict the forthcoming faulty IoD. Thereafter, data routing is performed over the updated network topology. Furthermore, to improve the network lifetime, the node's degree and betweenness centrality measures-based energy allocation method is also proposed. Finally, numerical results on simulated and real-field testbeds demonstrate the ENADL method.s effectiveness in predicting faulty nodes and re-routing data packets. This results in maximized network throughput and lifetime as compared to several existing methods.
{"title":"ENADL: Towards Performance Improvement of IoT Networks Using Deep Learning-Based Node Fault Prediction","authors":"Shraddha Tripathi;Faheem Nizar;Om Jee Pandey;Tushar Sandhan;Rajesh M. Hegde","doi":"10.1109/TR.2025.3540891","DOIUrl":"https://doi.org/10.1109/TR.2025.3540891","url":null,"abstract":"The Internet of Things (IoT) has grown explosively with wireless technology integration. Several IoT applications require high data throughput, low data transmission latency, and high data gathering reliability. Since, the IoT network (IoTN) is generally dynamic and utilizes a multi-hop data transmission scheme for such applications, the throughput, latency, and network lifetime tend to degrade as the hops increase. Moreover, IoT devices (IoD) are low-cost, less computationally capable, and battery-limited, further impacting performance. A faulty IoD worsens network lifetime and throughput. Predicting faulty nodes and re-routing data can significantly enhance performance. This work proposes a node fault prediction framework to enhance data routing in dynamic IoTN, maximizing throughput and lifetime. The network is represented as a graph in which the IoD are the nodes. Then a novel deep learning model is proposed utilizing various node and edge features to predict the faulty IoDs. Particularly, the proposed edge and node features-accumulation deep learning (ENADL) method exploits features, such as Euclidean distance between nodes, residual energy level of nodes, and type and number of messages passed between edges to predict the forthcoming faulty IoD. Thereafter, data routing is performed over the updated network topology. Furthermore, to improve the network lifetime, the node's degree and betweenness centrality measures-based energy allocation method is also proposed. Finally, numerical results on simulated and real-field testbeds demonstrate the ENADL method.s effectiveness in predicting faulty nodes and re-routing data packets. This results in maximized network throughput and lifetime as compared to several existing methods.","PeriodicalId":56305,"journal":{"name":"IEEE Transactions on Reliability","volume":"74 3","pages":"3514-3528"},"PeriodicalIF":5.7,"publicationDate":"2025-03-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144998138","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Man Ho Ling;Suk Joo Bae;Shengxin Jin;Hon Keung Tony Ng
Accelerated destructive degradation testing (ADDT) has become an invaluable method in reliability analysis, especially for highly reliable products. A common characteristic in many degradation studies is the presence of randomness in the initial degradation levels of testing units. Products with poor initial degradation levels tend to fail earlier. This study proposes an extended gamma process model that accommodates the random initial degradation value to accurately describe the degradation process over time. Under this modeling approach, we propose approximation methods for the conditional mean-time-to-failure (MTTF) and conditional variance of failure times to evaluate the impacts of initial degradation levels on product quality and reliability. We adopt a maximum likelihood approach to estimate the model parameters and MTTF under normal use conditions. In addition, we determine the optimal initial degradation threshold for removing poor-quality products and the proportion of products below this threshold. Based on the proposed model, the optimal ADDT plan is derived by minimizing the asymptotic variance of estimated MTTF under normal use conditions. A Monte Carlo simulation is conducted to assess the performance of the proposed inferential methods. Finally, a real-world ADDT dataset is analyzed to illustrate the proposed model and methodologies for making informed decisions on quality and reliability management.
{"title":"An Extended Gamma Process for Accelerated Destructive Degradation Test: Modeling and Optimal Design","authors":"Man Ho Ling;Suk Joo Bae;Shengxin Jin;Hon Keung Tony Ng","doi":"10.1109/TR.2025.3544545","DOIUrl":"https://doi.org/10.1109/TR.2025.3544545","url":null,"abstract":"Accelerated destructive degradation testing (ADDT) has become an invaluable method in reliability analysis, especially for highly reliable products. A common characteristic in many degradation studies is the presence of randomness in the initial degradation levels of testing units. Products with poor initial degradation levels tend to fail earlier. This study proposes an extended gamma process model that accommodates the random initial degradation value to accurately describe the degradation process over time. Under this modeling approach, we propose approximation methods for the conditional mean-time-to-failure (MTTF) and conditional variance of failure times to evaluate the impacts of initial degradation levels on product quality and reliability. We adopt a maximum likelihood approach to estimate the model parameters and MTTF under normal use conditions. In addition, we determine the optimal initial degradation threshold for removing poor-quality products and the proportion of products below this threshold. Based on the proposed model, the optimal ADDT plan is derived by minimizing the asymptotic variance of estimated MTTF under normal use conditions. A Monte Carlo simulation is conducted to assess the performance of the proposed inferential methods. Finally, a real-world ADDT dataset is analyzed to illustrate the proposed model and methodologies for making informed decisions on quality and reliability management.","PeriodicalId":56305,"journal":{"name":"IEEE Transactions on Reliability","volume":"74 3","pages":"4387-4401"},"PeriodicalIF":5.7,"publicationDate":"2025-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144998381","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Editorial: Applied AI for Reliability and Cybersecurity","authors":"Winston Shieh","doi":"10.1109/TR.2025.3541482","DOIUrl":"https://doi.org/10.1109/TR.2025.3541482","url":null,"abstract":"","PeriodicalId":56305,"journal":{"name":"IEEE Transactions on Reliability","volume":"74 1","pages":"1996-1997"},"PeriodicalIF":5.0,"publicationDate":"2025-03-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10910007","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143553333","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：