Reliability Engineering & System Safety最新文献_第9页

A stage-dependent Markov-switching fractional Brownian motion model for reliability analysis considering random effects and long-term memory 考虑随机效应和长期记忆的可靠性分析阶段相关马尔可夫开关分数布朗运动模型

IF 11 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Reliability Engineering & System Safety

Pub Date : 2026-01-07 DOI: 10.1016/j.ress.2026.112208

Yaohui Lu , Chao Zhang , Shaoping Wang , Rentong Chen , Shaoyang Du , Rui Mu , Dariusz Mazurkiewicz

Reliability management is crucial for ensuring stable operation of mechatronics components, as well as reducing the downtime and the operating costs. However, the existing degradation models based on Markov properties are not applicable because of the long-term memory of the components. In addition, the degradation of many components in their life cycle exhibits multi-stages, and dependencies exist between different degradation stages. Therefore, this paper proposes a stage-dependent Markov-switching fractional Brownian motion (FBM) model allowing to better capture the characteristics of nonlinearity, randomness, unit-to-unit variability, long-term memory, and dependency of multi-stage degradation. More precisely, the long-term memory of degradation is represented by the FBM process, and random effects are used to describe the unit-to-unit variability. Moreover, a stage-dependent Markov-switching process is proposed for describing the state transitions of multi-stage degradation processes. The working conditions of the different degradation stages are then used to describe the stage impact levels. Furthermore, the unknown parameters of the Markov-switching process and the nonlinear degradation model with FBM are determined based on the two-stage parameter estimation method. Finally, a simulation study and a real case on hydraulic pumps are conducted to demonstrate the high performance of the proposed model.

可靠性管理对于确保机电一体化部件的稳定运行，减少停机时间和运行成本至关重要。然而，现有的基于马尔可夫特性的退化模型由于部件的长期记忆性而不适用。此外，许多部件在其生命周期内的降解表现为多阶段，并且不同降解阶段之间存在依赖关系。因此，本文提出了一种阶段依赖的马尔可夫开关分数布朗运动（FBM）模型，该模型可以更好地捕捉非线性、随机性、单位间可变性、长期记忆和多阶段退化依赖性的特征。更准确地说，退化的长期记忆由FBM过程表示，并使用随机效应来描述单位间的可变性。此外，提出了一种阶段相关的马尔可夫切换过程来描述多阶段退化过程的状态转换。然后用不同降解阶段的工作条件来描述阶段影响水平。此外，基于两阶段参数估计方法确定了马尔可夫切换过程的未知参数和FBM非线性退化模型。最后，通过液压泵的仿真研究和实例验证了该模型的有效性。

{"title":"A stage-dependent Markov-switching fractional Brownian motion model for reliability analysis considering random effects and long-term memory","authors":"Yaohui Lu , Chao Zhang , Shaoping Wang , Rentong Chen , Shaoyang Du , Rui Mu , Dariusz Mazurkiewicz","doi":"10.1016/j.ress.2026.112208","DOIUrl":"10.1016/j.ress.2026.112208","url":null,"abstract":"<div><div>Reliability management is crucial for ensuring stable operation of mechatronics components, as well as reducing the downtime and the operating costs. However, the existing degradation models based on Markov properties are not applicable because of the long-term memory of the components. In addition, the degradation of many components in their life cycle exhibits multi-stages, and dependencies exist between different degradation stages. Therefore, this paper proposes a stage-dependent Markov-switching fractional Brownian motion (FBM) model allowing to better capture the characteristics of nonlinearity, randomness, unit-to-unit variability, long-term memory, and dependency of multi-stage degradation. More precisely, the long-term memory of degradation is represented by the FBM process, and random effects are used to describe the unit-to-unit variability. Moreover, a stage-dependent Markov-switching process is proposed for describing the state transitions of multi-stage degradation processes. The working conditions of the different degradation stages are then used to describe the stage impact levels. Furthermore, the unknown parameters of the Markov-switching process and the nonlinear degradation model with FBM are determined based on the two-stage parameter estimation method. Finally, a simulation study and a real case on hydraulic pumps are conducted to demonstrate the high performance of the proposed model.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"271 ","pages":"Article 112208"},"PeriodicalIF":11.0,"publicationDate":"2026-01-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146039665","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Extreme buffeting response of long-span bridges under probabilistic wind field: Environmental contours vs. brute-force Monte Carlo approaches 概率风场下大跨度桥梁的极端抖振响应：环境轮廓与蛮力蒙特卡罗方法

IF 11 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Reliability Engineering & System Safety

Pub Date : 2026-01-07 DOI: 10.1016/j.ress.2026.112199

Zihang Liu , Genshen Fang , Lin Zhao , Yaojun Ge , Nikolaos Nikitas , Enrico Zio

Turbulence parameters, which exhibit substantial uncertainty, are often disregarded in extreme buffeting evaluations. This study examines the long-term extreme buffeting response of a long-span bridge under a probabilistic wind field. First, the mean and turbulent wind field is analyzed to construct wind environmental contours. Then, a machine learning-based surrogate model using eXtreme Gradient Boosting (XGBoost) is employed to efficiently predict buffeting responses while reducing computational costs. Two strategies for computing long-term extreme buffeting responses are examined: (1) response evaluation along wind environmental contours and (2) direct estimation of annual extreme responses using Monte Carlo simulation (MCS) and the surrogate model. Results demonstrate that turbulence parameter uncertainty has a significant impact on the buffeting responses of the Xihoumen Bridge, with maximum torsional and vertical responses occurring under different wind conditions. Moreover, long-term extreme wind environment parameters do not always correspond to long-term extreme structural responses, underscoring the necessity of incorporating multiple turbulence parameters to accurately characterize wind-induced effects. The environmental contour method offers an effective hazard-oriented design strategy, and future work could explore the response-oriented design approaches that directly target structural performance.

湍流参数具有很大的不确定性，在极端抖振评估中经常被忽略。本文研究了大跨度桥梁在概率风场作用下的长期极端抖振响应。首先对平均风场和湍流风场进行分析，构建风环境等高线；然后，采用基于机器学习的代理模型，使用极限梯度增强（XGBoost）来有效预测抖振响应，同时降低计算成本。研究了计算长期极端抖振响应的两种策略：(1)沿风环境等高线的响应评估；(2)使用蒙特卡罗模拟（MCS）和代理模型直接估计年极端响应。结果表明，湍流参数的不确定性对西堠门大桥的抖振响应有显著影响，在不同的风况下，桥体的扭转和竖向响应均达到最大。此外，长期极端风环境参数并不总是与长期极端结构响应相对应，这强调了整合多个湍流参数以准确表征风致效应的必要性。环境轮廓法提供了一种有效的面向危害的设计策略，未来的工作可以探索直接针对结构性能的面向响应的设计方法。

{"title":"Extreme buffeting response of long-span bridges under probabilistic wind field: Environmental contours vs. brute-force Monte Carlo approaches","authors":"Zihang Liu , Genshen Fang , Lin Zhao , Yaojun Ge , Nikolaos Nikitas , Enrico Zio","doi":"10.1016/j.ress.2026.112199","DOIUrl":"10.1016/j.ress.2026.112199","url":null,"abstract":"<div><div>Turbulence parameters, which exhibit substantial uncertainty, are often disregarded in extreme buffeting evaluations. This study examines the long-term extreme buffeting response of a long-span bridge under a probabilistic wind field. First, the mean and turbulent wind field is analyzed to construct wind environmental contours. Then, a machine learning-based surrogate model using eXtreme Gradient Boosting (XGBoost) is employed to efficiently predict buffeting responses while reducing computational costs. Two strategies for computing long-term extreme buffeting responses are examined: (1) response evaluation along wind environmental contours and (2) direct estimation of annual extreme responses using Monte Carlo simulation (MCS) and the surrogate model. Results demonstrate that turbulence parameter uncertainty has a significant impact on the buffeting responses of the Xihoumen Bridge, with maximum torsional and vertical responses occurring under different wind conditions. Moreover, long-term extreme wind environment parameters do not always correspond to long-term extreme structural responses, underscoring the necessity of incorporating multiple turbulence parameters to accurately characterize wind-induced effects. The environmental contour method offers an effective hazard-oriented design strategy, and future work could explore the response-oriented design approaches that directly target structural performance.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"270 ","pages":"Article 112199"},"PeriodicalIF":11.0,"publicationDate":"2026-01-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145978683","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Identification and control of core causes of urban systemic risks under extreme precipitation 极端降水条件下城市系统性风险核心成因识别与控制

IF 11 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Reliability Engineering & System Safety

Pub Date : 2026-01-07 DOI: 10.1016/j.ress.2026.112210

Yu Lei , Shan Lu , Xiaojun Jiang

Extreme precipitation is highly susceptible to serious consequences in terms of urban systemic risk. Identifying and managing the core causes of urban systemic risks is the key to effective governance. In order to accurately identify and manage the core causes of urban systemic risks, this paper proposes a core cause analysis method for urban systemic risks under extreme precipitation using social network analysis. The method is verified by the “July 20″ heavy rainstorm disaster in Zhengzhou, China. The following conclusions are drawn: (1) Combining the intersection of degree centrality, collinearity frequency and cluster analysis, the core causes are divided into three categories: center core, important core and common core causes. (2) The two center core causes are failure of warning preparation and failure of information reporting. The two important core causes are engineering maintenance hazards and failure of response activation. The common core causes have 9 causes. (3) There are some evidences that core causes of urban systemic risks are dynamically heterogeneous in different stages extreme precipitation. (4) Removing the center core causes and weakening the important core causes can reduce the density of the cause network by 22% and increase the average path length by 20%, thereby effectively controlling urban systemic risks.

极端降水在城市系统性风险方面极易产生严重后果。识别和管理城市系统性风险的核心原因是有效治理的关键。为了准确识别和管理城市系统性风险的核心原因，本文提出了一种基于社会网络分析的极端降水条件下城市系统性风险核心原因分析方法。该方法以“7·20″”中国郑州特大暴雨灾害为例进行了验证。得出以下结论：(1)结合度中心性、共线性频率和聚类分析的交集，将核心原因分为中心核心、重要核心和共同核心三大类。(2)两个中心核心原因是预警准备失败和信息报告失败。两个重要的核心原因是工程维修危害和响应激活失效。常见的核心原因有9个。(3)不同极端降水阶段城市系统性风险的核心成因存在动态异质性。④去除中心核心原因，弱化重要核心原因，可使原因网络密度降低22%，平均路径长度增加20%，从而有效控制城市系统性风险。

{"title":"Identification and control of core causes of urban systemic risks under extreme precipitation","authors":"Yu Lei , Shan Lu , Xiaojun Jiang","doi":"10.1016/j.ress.2026.112210","DOIUrl":"10.1016/j.ress.2026.112210","url":null,"abstract":"<div><div>Extreme precipitation is highly susceptible to serious consequences in terms of urban systemic risk. Identifying and managing the core causes of urban systemic risks is the key to effective governance. In order to accurately identify and manage the core causes of urban systemic risks, this paper proposes a core cause analysis method for urban systemic risks under extreme precipitation using social network analysis. The method is verified by the “July 20″ heavy rainstorm disaster in Zhengzhou, China. The following conclusions are drawn: (1) Combining the intersection of degree centrality, collinearity frequency and cluster analysis, the core causes are divided into three categories: center core, important core and common core causes. (2) The two center core causes are failure of warning preparation and failure of information reporting. The two important core causes are engineering maintenance hazards and failure of response activation. The common core causes have 9 causes. (3) There are some evidences that core causes of urban systemic risks are dynamically heterogeneous in different stages extreme precipitation. (4) Removing the center core causes and weakening the important core causes can reduce the density of the cause network by 22% and increase the average path length by 20%, thereby effectively controlling urban systemic risks.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"270 ","pages":"Article 112210"},"PeriodicalIF":11.0,"publicationDate":"2026-01-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145978776","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Quantification of common cause failures with reduced order models using GO-FLOW method 用GO-FLOW方法用降阶模型量化共因故障

IF 11 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Reliability Engineering & System Safety

Pub Date : 2026-01-06 DOI: 10.1016/j.ress.2026.112197

Chuanlong Jiang , Fengjun Li , Ming Wang , Jun Yang

Common cause failures (CCFs) are critical factors in compromising reliability performance of safety-related systems due to their potential to cause multiple components to fail simultaneously under a shared cause. Traditional CCF analysis methods often rely on static group assumptions and fail to account for dynamic adjustments driven by technological advances and updates to operational data. In the paper, a novel algorithm is proposed for CCF analysis with reduced-order models using GO-FLOW method. The superiority and effectiveness of method is demonstrated with a benchmark case of auxiliary feedwater system (AFW) leakage failure in nuclear power plants (NPPs) between GO-FLOW and fault tree analysis (FTA). The conditioning of common cause failure probabilities for the common cause component group (CCCG) associated with observed component failure and planned maintenance and testing is also examined in the comparative study among the

β, α

, and MGL models. The results demonstrate that dynamic configuration changes within a CCCG have a pronounced impact on common cause failure. Exact GO-FLOW solutions can be achieved with the incorporation of various parametric models for both quantitative and qualitative analysis of common cause failures.

共因故障（CCFs）是影响安全相关系统可靠性性能的关键因素，因为它们有可能导致多个部件在共同原因下同时失效。传统的CCF分析方法通常依赖于静态的群体假设，无法解释由技术进步和操作数据更新驱动的动态调整。本文提出了一种基于GO-FLOW方法的CCF降阶分析算法。以核电厂辅助给水系统（AFW）泄漏事故为例，验证了该方法在GO-FLOW和故障树分析之间的优越性和有效性。在β、α和MGL模型的比较研究中，还考察了与观察到的部件故障和计划维护和测试相关的共因部件组（CCCG）的共因故障概率的条件。结果表明，CCCG内的动态配置变化对共因故障有显著的影响。通过结合各种参数模型对共因故障进行定量和定性分析，可以获得精确的GO-FLOW解决方案。

{"title":"Quantification of common cause failures with reduced order models using GO-FLOW method","authors":"Chuanlong Jiang , Fengjun Li , Ming Wang , Jun Yang","doi":"10.1016/j.ress.2026.112197","DOIUrl":"10.1016/j.ress.2026.112197","url":null,"abstract":"<div><div>Common cause failures (CCFs) are critical factors in compromising reliability performance of safety-related systems due to their potential to cause multiple components to fail simultaneously under a shared cause. Traditional CCF analysis methods often rely on static group assumptions and fail to account for dynamic adjustments driven by technological advances and updates to operational data. In the paper, a novel algorithm is proposed for CCF analysis with reduced-order models using GO-FLOW method. The superiority and effectiveness of method is demonstrated with a benchmark case of auxiliary feedwater system (AFW) leakage failure in nuclear power plants (NPPs) between GO-FLOW and fault tree analysis (FTA). The conditioning of common cause failure probabilities for the common cause component group (CCCG) associated with observed component failure and planned maintenance and testing is also examined in the comparative study among the <span><math><mrow><mi>β</mi><mo>,</mo><mspace></mspace><mi>α</mi></mrow></math></span>, and MGL models. The results demonstrate that dynamic configuration changes within a CCCG have a pronounced impact on common cause failure. Exact GO-FLOW solutions can be achieved with the incorporation of various parametric models for both quantitative and qualitative analysis of common cause failures.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"270 ","pages":"Article 112197"},"PeriodicalIF":11.0,"publicationDate":"2026-01-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145978773","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Collaborative domain adaptation-based bidirectional temporal convolution network for cross-condition and cross-equipment remaining useful life prediction of rolling bearings 基于协同域自适应的双向时间卷积网络用于滚动轴承跨工况、跨设备剩余使用寿命预测

IF 11 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Reliability Engineering & System Safety

Pub Date : 2026-01-06 DOI: 10.1016/j.ress.2026.112193

Feiyue Deng , Rujiang Hao , Shaopu Yang , Boxun Luo , Jin Sha

Accurate prediction of the remaining useful life (RUL) of rolling bearings is of paramount importance in industrial settings. Nevertheless, traditional deep learning predictive models face challenges in handling complex, multi-condition bearing data and often inadequate generalization capabilities across varying operational scenarios. Existing domain adaptive (DA)-based cross-domain RUL prediction methods suffer from fuzzy decision boundaries and poor comprehensive feature recognition. This study proposes a collaborative DA-based bidirectional temporal convolution network (CDBTCN) for cross-condition and cross-equipment RUL prediction of rolling bearings. The model employs an innovative parallel bidirectional architecture to simultaneously process forward and reverse time series, enhancing temporal feature extraction and reducing noise interference. By constructing two independent RUL predictor heads, the fuzziness of decision boundaries is alleviated by calculating the maximum prediction difference. This study further introduces an adaptive Wasserstein-integrated multi-kernel MMD (AWI-MK-MMD)-based strategy to minimize the distribution discrepancy of samples between the source and target domains. Extensive experiments demonstrate that the proposed method achieves more accurate RUL prediction results under cross-condition and cross-equipment scenarios, outperforming CNN-based, TCN-based, and traditional DA-based models.

准确预测滚动轴承的剩余使用寿命（RUL）在工业环境中至关重要。然而，传统的深度学习预测模型在处理复杂的、多条件的轴承数据方面面临挑战，并且在不同的操作场景中往往缺乏泛化能力。现有的基于领域自适应（DA）的跨领域规则语义预测方法存在决策边界模糊和综合特征识别能力差的问题。本研究提出了一种基于协同数据的双向时间卷积网络（CDBTCN），用于滚动轴承的跨条件和跨设备RUL预测。该模型采用创新的并行双向架构，同时对时间序列进行正反向处理，增强了时间特征提取能力，降低了噪声干扰。通过构造两个独立的RUL预测头，通过计算最大预测差来缓解决策边界的模糊性。本研究进一步引入了一种基于自适应wasserstein集成多核MMD （AWI-MK-MMD）的策略，以最小化源域和目标域之间的样本分布差异。大量实验表明，该方法在跨条件和跨设备场景下获得了更准确的RUL预测结果，优于基于cnn、基于tcn和传统的基于数据的模型。

{"title":"Collaborative domain adaptation-based bidirectional temporal convolution network for cross-condition and cross-equipment remaining useful life prediction of rolling bearings","authors":"Feiyue Deng , Rujiang Hao , Shaopu Yang , Boxun Luo , Jin Sha","doi":"10.1016/j.ress.2026.112193","DOIUrl":"10.1016/j.ress.2026.112193","url":null,"abstract":"<div><div>Accurate prediction of the remaining useful life (RUL) of rolling bearings is of paramount importance in industrial settings. Nevertheless, traditional deep learning predictive models face challenges in handling complex, multi-condition bearing data and often inadequate generalization capabilities across varying operational scenarios. Existing domain adaptive (DA)-based cross-domain RUL prediction methods suffer from fuzzy decision boundaries and poor comprehensive feature recognition. This study proposes a collaborative DA-based bidirectional temporal convolution network (CDBTCN) for cross-condition and cross-equipment RUL prediction of rolling bearings. The model employs an innovative parallel bidirectional architecture to simultaneously process forward and reverse time series, enhancing temporal feature extraction and reducing noise interference. By constructing two independent RUL predictor heads, the fuzziness of decision boundaries is alleviated by calculating the maximum prediction difference. This study further introduces an adaptive Wasserstein-integrated multi-kernel MMD (AWI-MK-MMD)-based strategy to minimize the distribution discrepancy of samples between the source and target domains. Extensive experiments demonstrate that the proposed method achieves more accurate RUL prediction results under cross-condition and cross-equipment scenarios, outperforming CNN-based, TCN-based, and traditional DA-based models.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"270 ","pages":"Article 112193"},"PeriodicalIF":11.0,"publicationDate":"2026-01-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145927923","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Expected utility-based maintenance planning under risk aversion for a two-state deterioration model 风险规避下基于预期效用的双状态劣化模型维修计划

IF 11 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Reliability Engineering & System Safety

Pub Date : 2026-01-06 DOI: 10.1016/j.ress.2025.112169

Niyazi Onur Bakır , Salih Tekin , Büşra Keleş

In this study, we present a risk-averse maintenance decision model that is based on a binary state formulation subject to non-self-announcing failures where a system breakdown is detected only through a periodic inspection. We generalize the fundamental binary state formulation to permit gradual system aging and analyze how the system intervention decision changes in response to shifts in the degree of risk aversion. The maintenance decision problem is modeled within a risk-averse expected-utility framework and solved using renewal theory. Our findings suggest that risky intervention alternatives are less preferable under higher risk aversion, especially when parameters related to the system downtime are significantly high. Under some circumstances, we also observe that it may become uneconomic to continue business for a risk-averse system administrator, although the risk-neutral decision is to remain operational. The results also show that the sensitivity of maintenance decisions to risk preferences increases as the model is adjusted to further highlight the trade-off between risky repair and risk-free replacement alternatives.

在本研究中，我们提出了一种规避风险的维护决策模型，该模型基于非自我宣布故障的二元状态公式，其中只有通过定期检查才能检测到系统故障。我们推广了基本的二元状态公式以允许系统逐渐老化，并分析了系统干预决策如何随着风险厌恶程度的变化而变化。在规避风险的预期效用框架中对维护决策问题进行建模，并使用更新理论进行求解。我们的研究结果表明，在风险厌恶程度较高的情况下，风险干预方案不太可取，特别是当与系统停机时间相关的参数显著较高时。在某些情况下，我们还观察到，尽管风险中性的决策是保持可操作性，但对于厌恶风险的系统管理员来说，继续业务可能变得不经济。结果还表明，维修决策对风险偏好的敏感性随着模型的调整而增加，以进一步突出有风险的维修和无风险的更换方案之间的权衡。

{"title":"Expected utility-based maintenance planning under risk aversion for a two-state deterioration model","authors":"Niyazi Onur Bakır , Salih Tekin , Büşra Keleş","doi":"10.1016/j.ress.2025.112169","DOIUrl":"10.1016/j.ress.2025.112169","url":null,"abstract":"<div><div>In this study, we present a risk-averse maintenance decision model that is based on a binary state formulation subject to non-self-announcing failures where a system breakdown is detected only through a periodic inspection. We generalize the fundamental binary state formulation to permit gradual system aging and analyze how the system intervention decision changes in response to shifts in the degree of risk aversion. The maintenance decision problem is modeled within a risk-averse expected-utility framework and solved using renewal theory. Our findings suggest that risky intervention alternatives are less preferable under higher risk aversion, especially when parameters related to the system downtime are significantly high. Under some circumstances, we also observe that it may become uneconomic to continue business for a risk-averse system administrator, although the risk-neutral decision is to remain operational. The results also show that the sensitivity of maintenance decisions to risk preferences increases as the model is adjusted to further highlight the trade-off between risky repair and risk-free replacement alternatives.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"270 ","pages":"Article 112169"},"PeriodicalIF":11.0,"publicationDate":"2026-01-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145928526","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Sensitivity estimation of stochastic output with respect to distribution parameters of stochastic inputs 随机输出对随机输入分布参数的灵敏度估计

IF 11 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Reliability Engineering & System Safety

Pub Date : 2026-01-06 DOI: 10.1016/j.ress.2026.112191

Xuan-Yi Zhang , Yan-Gang Zhao , Marcos A. Valdebenito , Matthias G.R. Faes

Computational models have become indispensable tools for decision-making across numerous fields. Given the inherent randomness in input variables, the outputs of these models are often stochastic, making sensitivity estimation (SE) essential for understanding how variations in inputs affect stochastic outputs. In practice, the input random variables are described by their distribution parameters. This study introduces an SE method to assess the influence of input distribution parameters on the moments and distributions of outputs. Sensitivity indices (SIs) are defined based on both the first three moments and the cumulative distribution function of the outputs, naturally providing SI for exceeding probabilities. A numerical approach is developed to quantify these SIs as the post processing of uncertainty quantification, employing a moment-based model to approximate the output distribution. Three examples, including nonlinear formula and finite element model, are analyzed to demonstrate the applicability and efficiency of the proposed SE method, highlighting its ability to provide a more comprehensive view of the relationship between input distribution parameters and model outputs.

计算模型已经成为许多领域决策不可或缺的工具。考虑到输入变量固有的随机性，这些模型的输出通常是随机的，因此灵敏度估计（SE）对于理解输入变量如何影响随机输出至关重要。在实际中，输入的随机变量是由它们的分布参数来描述的。本文介绍了一种SE方法来评估输入分布参数对输出矩和分布的影响。灵敏度指数（SI）是根据前三个矩和输出的累积分布函数定义的，自然地提供了超过概率的SI。采用基于矩的模型来近似输出分布，开发了一种数值方法来量化这些si作为不确定性量化的后处理。通过对非线性公式和有限元模型三个实例的分析，证明了所提出的SE方法的适用性和有效性，突出了其能够更全面地了解输入分布参数与模型输出之间的关系。

{"title":"Sensitivity estimation of stochastic output with respect to distribution parameters of stochastic inputs","authors":"Xuan-Yi Zhang , Yan-Gang Zhao , Marcos A. Valdebenito , Matthias G.R. Faes","doi":"10.1016/j.ress.2026.112191","DOIUrl":"10.1016/j.ress.2026.112191","url":null,"abstract":"<div><div>Computational models have become indispensable tools for decision-making across numerous fields. Given the inherent randomness in input variables, the outputs of these models are often stochastic, making sensitivity estimation (SE) essential for understanding how variations in inputs affect stochastic outputs. In practice, the input random variables are described by their distribution parameters. This study introduces an SE method to assess the influence of input distribution parameters on the moments and distributions of outputs. Sensitivity indices (SIs) are defined based on both the first three moments and the cumulative distribution function of the outputs, naturally providing SI for exceeding probabilities. A numerical approach is developed to quantify these SIs as the post processing of uncertainty quantification, employing a moment-based model to approximate the output distribution. Three examples, including nonlinear formula and finite element model, are analyzed to demonstrate the applicability and efficiency of the proposed SE method, highlighting its ability to provide a more comprehensive view of the relationship between input distribution parameters and model outputs.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"270 ","pages":"Article 112191"},"PeriodicalIF":11.0,"publicationDate":"2026-01-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145978780","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Reliability-Based Maintenance Optimization of Reusable Phased Mission Systems 基于可靠性的可重复使用分阶段任务系统维护优化

IF 11 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Reliability Engineering & System Safety

Pub Date : 2026-01-06 DOI: 10.1016/j.ress.2026.112187

Xiang-Yu Li , Haochen Wang , He Li , Xiaoyan Xiong , Zaili Yang , Hong-Zhong Huang , Michael Beer , Jin Wang

This paper proposes a reliability-based maintenance optimization method for reliability modeling and maintenance planning of reusable phased mission systems (R-PMSs). Initially, a multi-phased Wiener process-based reliability modeling method, combined with the Binary Decision Diagram-based model, is created to assess the reliability of R-PMSs. The method considers the initial states of components to be perfect/imperfect fixed as a basis to reflect the impacts of multiple usages of R-PMSs. Subsequently, a maintenance optimization model based on the reliability assessed and incorporating multiple maintenance actions is constructed to allocate the maintenance resources into the maintenance activities in advance of subsequent missions of R-PMSs. The model balances the maintenance cost, overall reliability, and task requirements. The superiority and performance of the proposed method are validated by numerical and engineering analysis. The results confirmed that the proposed method contributes to reliability modeling and maintenance scheduling of R-PMSs.

针对可重用分阶段任务系统的可靠性建模和维修规划问题，提出了一种基于可靠性的维修优化方法。首先，建立了一种基于Wiener过程的多阶段可靠性建模方法，并结合基于二元决策图的模型对r - pms进行可靠性评估。该方法以组分的初始状态完全固定/不完全固定为基础，反映了r - pms多种使用的影响。在此基础上，构建了基于可靠性评估、包含多个维修动作的维修优化模型，在r - pms后续任务前将维修资源分配到维修活动中。该模型平衡了维护成本、总体可靠性和任务需求。数值分析和工程分析验证了该方法的优越性和性能。结果表明，该方法有助于r - pms的可靠性建模和维修调度。

{"title":"Reliability-Based Maintenance Optimization of Reusable Phased Mission Systems","authors":"Xiang-Yu Li , Haochen Wang , He Li , Xiaoyan Xiong , Zaili Yang , Hong-Zhong Huang , Michael Beer , Jin Wang","doi":"10.1016/j.ress.2026.112187","DOIUrl":"10.1016/j.ress.2026.112187","url":null,"abstract":"<div><div>This paper proposes a reliability-based maintenance optimization method for reliability modeling and maintenance planning of reusable phased mission systems (R-PMSs). Initially, a multi-phased Wiener process-based reliability modeling method, combined with the Binary Decision Diagram-based model, is created to assess the reliability of R-PMSs. The method considers the initial states of components to be perfect/imperfect fixed as a basis to reflect the impacts of multiple usages of R-PMSs. Subsequently, a maintenance optimization model based on the reliability assessed and incorporating multiple maintenance actions is constructed to allocate the maintenance resources into the maintenance activities in advance of subsequent missions of R-PMSs. The model balances the maintenance cost, overall reliability, and task requirements. The superiority and performance of the proposed method are validated by numerical and engineering analysis. The results confirmed that the proposed method contributes to reliability modeling and maintenance scheduling of R-PMSs.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"270 ","pages":"Article 112187"},"PeriodicalIF":11.0,"publicationDate":"2026-01-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145978681","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

An active learning multi-fidelity Kriging model for predicting the expected lifetime of time- and space-dependent structural systems with multi-fidelity data 基于多保真度数据预测时空相关结构系统预期寿命的主动学习多保真度Kriging模型

IF 11 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Reliability Engineering & System Safety

Pub Date : 2026-01-06 DOI: 10.1016/j.ress.2026.112192

Hongyou Zhan , Hui Liu , Ning-Cong Xiao

Predicting the expected lifetime is a critical metric for quantifying uncertainties during the design phase. However, it typically requires numerous computationally expensive simulations. This study proposes a regression-based multi-fidelity Kriging model to predict the expected lifetime of time- and space-dependent systems with multi-fidelity data. The model integrates high-, medium-, and low-fidelity data, where the high-fidelity model is represented as a weighted combination of the lower-fidelity models with a discrepancy model. An initial multi-fidelity Kriging model is constructed with a limited number of training samples from different fidelity levels. To enhance model accuracy, a novel learning function is proposed to adaptively select training samples by considering random variables, time and space variables, and fidelity levels. The refinement process terminates when the relative error between the predicted and expected lifetimes falls below 0.001, thereby ensuring lifetime prediction accuracy. Three numerical examples are presented to validate the proposed method, demonstrating its efficiency in predicting the expected lifetime of structural systems involving nonlinearity, random fields, and implicit performance functions.

在设计阶段，预测预期寿命是量化不确定性的关键指标。然而，它通常需要大量计算上昂贵的模拟。本文提出了一种基于回归的多保真度Kriging模型来预测具有多保真度数据的时空依赖系统的预期寿命。该模型集成了高保真度、中保真度和低保真度数据，其中高保真度模型表示为带有差异模型的低保真度模型的加权组合。用有限数量的不同保真度的训练样本构建初始多保真度Kriging模型。为了提高模型的准确性，提出了一种新的学习函数，通过考虑随机变量、时间和空间变量以及保真度来自适应地选择训练样本。当预测寿命和预期寿命之间的相对误差低于0.001时，精化过程终止，从而确保寿命预测的准确性。最后给出了三个数值算例，验证了该方法在非线性、随机场和隐式性能函数的结构系统预期寿命预测中的有效性。

{"title":"An active learning multi-fidelity Kriging model for predicting the expected lifetime of time- and space-dependent structural systems with multi-fidelity data","authors":"Hongyou Zhan , Hui Liu , Ning-Cong Xiao","doi":"10.1016/j.ress.2026.112192","DOIUrl":"10.1016/j.ress.2026.112192","url":null,"abstract":"<div><div>Predicting the expected lifetime is a critical metric for quantifying uncertainties during the design phase. However, it typically requires numerous computationally expensive simulations. This study proposes a regression-based multi-fidelity Kriging model to predict the expected lifetime of time- and space-dependent systems with multi-fidelity data. The model integrates high-, medium-, and low-fidelity data, where the high-fidelity model is represented as a weighted combination of the lower-fidelity models with a discrepancy model. An initial multi-fidelity Kriging model is constructed with a limited number of training samples from different fidelity levels. To enhance model accuracy, a novel learning function is proposed to adaptively select training samples by considering random variables, time and space variables, and fidelity levels. The refinement process terminates when the relative error between the predicted and expected lifetimes falls below 0.001, thereby ensuring lifetime prediction accuracy. Three numerical examples are presented to validate the proposed method, demonstrating its efficiency in predicting the expected lifetime of structural systems involving nonlinearity, random fields, and implicit performance functions.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"270 ","pages":"Article 112192"},"PeriodicalIF":11.0,"publicationDate":"2026-01-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145978682","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Time-dependent periodic trajectory accuracy reliability analysis of redundant planar manipulator considering eigenvalue sensitivity analysis method 基于特征值灵敏度分析法的冗余平面机械臂时变周期轨迹精度可靠性分析

IF 11 1区工程技术 Q1 ENGINEERING, INDUSTRIAL

Reliability Engineering & System Safety

Pub Date : 2026-01-05 DOI: 10.1016/j.ress.2026.112194

Luchuan Yu, Wangqifei Shen, Shenquan Huang

For redundant robotic manipulators in automotive stamping lines, the reliability of periodic trajectory accuracy is crucial for production quality and efficiency. However, the cumulative effects of dynamic periodicity under complex working conditions are often overlooked by existing methods. To address this issue, this paper proposes an eigenvalue sensitivity analysis method for time-dependent periodic trajectory accuracy reliability analysis and enhancement of the 2-redundancy planar manipulator. First, A reliable trajectory generation model combining Fourier transform with Taylor expansion is constructed, enabling the delineation of key error propagation paths from Cartesian to joint space. Building on this foundation, an interpretable modeling-quantification-allocation trajectory correction framework is developed, where error propagation heatmaps are utilized to identify high-contributing joints and a dynamic scaling matrix is incorporated to achieve precise localization and efficient compensation of end-effector errors. To validate the effectiveness and generality of the proposed method, systematic tests are conducted across three representative workspace scales: compact, medium, and large. Through comparative analysis with Monte Carlo Simulation (MCS) experiments, the proposed method is demonstrated to consistently enhance trajectory accuracy across all tested workspaces, providing a novel and robust solution for high-precision motion reliability in industrial robotics.

对于汽车冲压生产线中的冗余机器人，周期轨迹精度的可靠性对生产质量和效率至关重要。然而，现有方法往往忽略了复杂工况下动态周期性的累积效应。针对这一问题，本文提出了一种特征值灵敏度分析方法，用于二冗余平面机械臂的时变周期轨迹精度、可靠性分析和增强。首先，建立了可靠的傅立叶变换与泰勒展开相结合的轨迹生成模型，实现了从笛卡尔空间到关节空间关键误差传播路径的描绘；在此基础上，建立了可解释的建模-量化-分配轨迹校正框架，利用误差传播热图识别高贡献关节，并结合动态缩放矩阵实现末端执行器误差的精确定位和有效补偿。为了验证所提出方法的有效性和通用性，在三个代表性的工作空间尺度上进行了系统的测试：紧凑型、中型和大型。通过与蒙特卡罗仿真（MCS）实验的对比分析，证明了该方法在所有测试工作空间中都能持续提高轨迹精度，为工业机器人高精度运动可靠性提供了一种新颖而稳健的解决方案。

{"title":"Time-dependent periodic trajectory accuracy reliability analysis of redundant planar manipulator considering eigenvalue sensitivity analysis method","authors":"Luchuan Yu, Wangqifei Shen, Shenquan Huang","doi":"10.1016/j.ress.2026.112194","DOIUrl":"10.1016/j.ress.2026.112194","url":null,"abstract":"<div><div>For redundant robotic manipulators in automotive stamping lines, the reliability of periodic trajectory accuracy is crucial for production quality and efficiency. However, the cumulative effects of dynamic periodicity under complex working conditions are often overlooked by existing methods. To address this issue, this paper proposes an eigenvalue sensitivity analysis method for time-dependent periodic trajectory accuracy reliability analysis and enhancement of the 2-redundancy planar manipulator. First, A reliable trajectory generation model combining Fourier transform with Taylor expansion is constructed, enabling the delineation of key error propagation paths from Cartesian to joint space. Building on this foundation, an interpretable modeling-quantification-allocation trajectory correction framework is developed, where error propagation heatmaps are utilized to identify high-contributing joints and a dynamic scaling matrix is incorporated to achieve precise localization and efficient compensation of end-effector errors. To validate the effectiveness and generality of the proposed method, systematic tests are conducted across three representative workspace scales: compact, medium, and large. Through comparative analysis with Monte Carlo Simulation (MCS) experiments, the proposed method is demonstrated to consistently enhance trajectory accuracy across all tested workspaces, providing a novel and robust solution for high-precision motion reliability in industrial robotics.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"270 ","pages":"Article 112194"},"PeriodicalIF":11.0,"publicationDate":"2026-01-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145928522","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0