Pub Date : 2026-01-23DOI: 10.1016/j.ress.2026.112280
Jiaxin Du , Yongtao Xi , Jinxian Weng , Bing Han , Haifeng Ding
Ensuring navigation safety is a key objective in maritime transport, particularly in restricted waters where human factors contribute predominantly to accidents. Complex operating conditions increase crew workload, induce physiological responses, and lead to ship behavioral changes that shape collision risk. Based on the Information-Decision-Action (IDA) theory, this study develops an Environment-Human state-Ship behavior-Consequence (EHSC) framework and constructs a real navigation data-driven Bayesian Network (BN). Real-world experiments on the Huangpu River were designed to investigate how environmental conditions influence seafarers’ states, which further affect ship behavior and risk. Results indicate that the minimum distance to other vessels and speed are the most sensitive determinants of collision risk. Low Galvanic Skin Response (GSR), which tends to occur under nighttime conditions, limited traffic interactions, or low traffic density, is associated with close-proximity navigation and sustained high speed. Captains aged 50–60 exhibit stronger risk management capabilities. These findings clarify human-performance pathways of collision risk and provide valuable support for early warning systems.
{"title":"Effects of human performance on ship collision risk in restricted waters: A Bayesian network driven by real navigation data","authors":"Jiaxin Du , Yongtao Xi , Jinxian Weng , Bing Han , Haifeng Ding","doi":"10.1016/j.ress.2026.112280","DOIUrl":"10.1016/j.ress.2026.112280","url":null,"abstract":"<div><div>Ensuring navigation safety is a key objective in maritime transport, particularly in restricted waters where human factors contribute predominantly to accidents. Complex operating conditions increase crew workload, induce physiological responses, and lead to ship behavioral changes that shape collision risk. Based on the Information-Decision-Action (IDA) theory, this study develops an Environment-Human state-Ship behavior-Consequence (EHSC) framework and constructs a real navigation data-driven Bayesian Network (BN). Real-world experiments on the Huangpu River were designed to investigate how environmental conditions influence seafarers’ states, which further affect ship behavior and risk. Results indicate that the minimum distance to other vessels and speed are the most sensitive determinants of collision risk. Low Galvanic Skin Response (GSR), which tends to occur under nighttime conditions, limited traffic interactions, or low traffic density, is associated with close-proximity navigation and sustained high speed. Captains aged 50–60 exhibit stronger risk management capabilities. These findings clarify human-performance pathways of collision risk and provide valuable support for early warning systems.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"271 ","pages":"Article 112280"},"PeriodicalIF":11.0,"publicationDate":"2026-01-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146079653","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}
Pub Date : 2026-01-23DOI: 10.1016/j.ress.2026.112271
Jiusi Zhang , Chunxiao Wang , Quan Qian , Shen Yin
As the complexity of industrial equipment continues to increase, determining the remaining useful life (RUL) with high precision holds substantial significance for maintaining intricate industrial systems. The development of cross-domain prognostic approaches without source domain data necessitates thorough investigation, given the inherent distribution shifts among edge devices’ degradation patterns and the imperative of preserving data security protocols. Furthermore, convolutional neural network, and long short-term memory network perform insufficiently when processing complex structurally dependent data. Consequently, this paper proposes a distributed RUL prediction approach based on graph convolutional neural network. Specifically, this paper designs a differential attention graph convolutional neural network that can focus on key areas in degradation data. Furthermore, considering the privacy and security of degradation data, this paper designs a two-stage decision boundary adjustment approach to achieve source-free RUL prediction under cross-domain conditions. On this basis, the study introduces a federated consensus mechanism that implements progressive weight calibration aligned with distributed training dynamics in edge computing environments, which can effectively reduce overfitting, and improve the generalization ability. Experimental validation on NASA’s publicly available aircraft engine degradation dataset confirms the operational efficacy of the proposed approach.
{"title":"Source-free domain adaptation for cross-domain remaining useful life prediction: A distributed federated learning perspective","authors":"Jiusi Zhang , Chunxiao Wang , Quan Qian , Shen Yin","doi":"10.1016/j.ress.2026.112271","DOIUrl":"10.1016/j.ress.2026.112271","url":null,"abstract":"<div><div>As the complexity of industrial equipment continues to increase, determining the remaining useful life (RUL) with high precision holds substantial significance for maintaining intricate industrial systems. The development of cross-domain prognostic approaches without source domain data necessitates thorough investigation, given the inherent distribution shifts among edge devices’ degradation patterns and the imperative of preserving data security protocols. Furthermore, convolutional neural network, and long short-term memory network perform insufficiently when processing complex structurally dependent data. Consequently, this paper proposes a distributed RUL prediction approach based on graph convolutional neural network. Specifically, this paper designs a differential attention graph convolutional neural network that can focus on key areas in degradation data. Furthermore, considering the privacy and security of degradation data, this paper designs a two-stage decision boundary adjustment approach to achieve source-free RUL prediction under cross-domain conditions. On this basis, the study introduces a federated consensus mechanism that implements progressive weight calibration aligned with distributed training dynamics in edge computing environments, which can effectively reduce overfitting, and improve the generalization ability. Experimental validation on NASA’s publicly available aircraft engine degradation dataset confirms the operational efficacy of the proposed approach.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"271 ","pages":"Article 112271"},"PeriodicalIF":11.0,"publicationDate":"2026-01-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146079475","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}
Pub Date : 2026-01-23DOI: 10.1016/j.ress.2026.112281
Qingwen Xiong , Xianbao Yuan , Sen Zhang , Jianjun Zhou , Zhangliang Mao , Yonghong Zhang
Model calibration is a technique that enhances computational accuracy by adjusting model inputs or structures, and can be categorized into probabilistic and non-probabilistic methods. In the field of nuclear reactors, limitations such as insufficient data, complex model structures, and numerous parameters often render probabilistic methods inapplicable in many scenarios. Meanwhile, non-probabilistic methods fail to account for model form uncertainty, making it difficult to accurately evaluate the confidence level and coverage. To address these challenges, a novel uncertainty informed calibration framework based on the non-probabilistic interval theory is proposed. The framework integrates techniques such as artificial neural networks, model uncertainty evaluation, double-loop nested sampling, and optimization algorithms, enabling the acquisition of non-probabilistic intervals for input parameters through inverse calibration. The proposed framework is validated using the critical flow model, and its reliability is verified by comparing the performance of multiple calibration methods. Subsequently, the framework is applied to the counter-current flow limitation model. The results demonstrate that the framework is suitable for inverse calibration even with limited observational data, as it accurately obtains input parameter intervals with a specific coverage rate (e.g., 95 %) while maintaining high computational efficiency.
{"title":"Uncertainty informed calibration of thermal-hydraulic models for nuclear reactor via integrated neural network and optimization algorithm framework","authors":"Qingwen Xiong , Xianbao Yuan , Sen Zhang , Jianjun Zhou , Zhangliang Mao , Yonghong Zhang","doi":"10.1016/j.ress.2026.112281","DOIUrl":"10.1016/j.ress.2026.112281","url":null,"abstract":"<div><div>Model calibration is a technique that enhances computational accuracy by adjusting model inputs or structures, and can be categorized into probabilistic and non-probabilistic methods. In the field of nuclear reactors, limitations such as insufficient data, complex model structures, and numerous parameters often render probabilistic methods inapplicable in many scenarios. Meanwhile, non-probabilistic methods fail to account for model form uncertainty, making it difficult to accurately evaluate the confidence level and coverage. To address these challenges, a novel uncertainty informed calibration framework based on the non-probabilistic interval theory is proposed. The framework integrates techniques such as artificial neural networks, model uncertainty evaluation, double-loop nested sampling, and optimization algorithms, enabling the acquisition of non-probabilistic intervals for input parameters through inverse calibration. The proposed framework is validated using the critical flow model, and its reliability is verified by comparing the performance of multiple calibration methods. Subsequently, the framework is applied to the counter-current flow limitation model. The results demonstrate that the framework is suitable for inverse calibration even with limited observational data, as it accurately obtains input parameter intervals with a specific coverage rate (e.g., 95 %) while maintaining high computational efficiency.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"271 ","pages":"Article 112281"},"PeriodicalIF":11.0,"publicationDate":"2026-01-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146079480","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}
Pub Date : 2026-01-23DOI: 10.1016/j.ress.2026.112282
Lichao Yang , Jingxian Liu , Zhao Liu , Qin Zhou , Yang Liu , Yukuan Wang , Weihuang Wu
To systematically investigate the complex causal mechanisms of maritime accidents, this study proposes an automated analytical framework that integrates Natural Language Processing (NLP) with complex network theory. The framework is designed to transform unstructured accident investigation reports into a quantifiable causal network that reflects systemic risk. Drawing on 564 official reports, this study constructs a standardised dataset of causal factors through a two-stage process combining automated preprocessing and manual coding. NLP techniques are then employed to extract causal relationships from the texts, enabling the construction of a weighted, directed complex network from discrete factors. To ensure the reliability of the framework, the extracted causal logic is verified by a domain expert panel, and the identified risk propagation patterns are validated against representative empirical cases. Topological analysis reveals that the causal network exhibits the “small-world” and “scale-free” properties characteristic of complex systems, indicating a high potential for efficient risk propagation mediated by a few key hubs. A multi-dimensional centrality assessment identifies static risk sources of high influence, including “Inadequate Supervision”, “Vessel Stability/Stowage Issues”, and “Adverse Weather/Sea State”. Furthermore, a risk pathway identification algorithm is applied to extract five typical risk propagation patterns. These pathways dynamically illustrate the systemic process by which risk evolves from latent managerial failures, through technical vulnerabilities and the actions of front-line personnel, to a major accident when triggered by specific environmental conditions. This work provides a dynamic, systematic network perspective for accident causation analysis, and its findings offer more precise intervention targets and process-based preventive strategies for maritime safety management.
{"title":"From text to network: A framework for identifying causal factors and risk propagation paths in maritime accidents","authors":"Lichao Yang , Jingxian Liu , Zhao Liu , Qin Zhou , Yang Liu , Yukuan Wang , Weihuang Wu","doi":"10.1016/j.ress.2026.112282","DOIUrl":"10.1016/j.ress.2026.112282","url":null,"abstract":"<div><div>To systematically investigate the complex causal mechanisms of maritime accidents, this study proposes an automated analytical framework that integrates Natural Language Processing (NLP) with complex network theory. The framework is designed to transform unstructured accident investigation reports into a quantifiable causal network that reflects systemic risk. Drawing on 564 official reports, this study constructs a standardised dataset of causal factors through a two-stage process combining automated preprocessing and manual coding. NLP techniques are then employed to extract causal relationships from the texts, enabling the construction of a weighted, directed complex network from discrete factors. To ensure the reliability of the framework, the extracted causal logic is verified by a domain expert panel, and the identified risk propagation patterns are validated against representative empirical cases. Topological analysis reveals that the causal network exhibits the “small-world” and “scale-free” properties characteristic of complex systems, indicating a high potential for efficient risk propagation mediated by a few key hubs. A multi-dimensional centrality assessment identifies static risk sources of high influence, including “Inadequate Supervision”, “Vessel Stability/Stowage Issues”, and “Adverse Weather/Sea State”. Furthermore, a risk pathway identification algorithm is applied to extract five typical risk propagation patterns. These pathways dynamically illustrate the systemic process by which risk evolves from latent managerial failures, through technical vulnerabilities and the actions of front-line personnel, to a major accident when triggered by specific environmental conditions. This work provides a dynamic, systematic network perspective for accident causation analysis, and its findings offer more precise intervention targets and process-based preventive strategies for maritime safety management.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"271 ","pages":"Article 112282"},"PeriodicalIF":11.0,"publicationDate":"2026-01-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146079554","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}
Pub Date : 2026-01-23DOI: 10.1016/j.ress.2026.112246
Yi Li , Fan Zhang , Jingzheng Liu , Faping Zhang , Tianci Liu , Fanyue Zhou
To address the challenges of anomaly monitoring arising from the increasing complexity and scale of Discrete Manufacturing Systems (DMS), this study proposes an anomaly monitoring method based on a Dual Joint Network (DJN), which integrates complex network theory with operational data. The model consists of two components: a real graph established through physical modeling and a virtual graph constructed through data-driven modeling. In the real graph, anomalies are detected by identifying abrupt changes in network topology relative to the Representative Graph (RG), which characterizes the normal operating state. In the virtual graph, an improved SpotLight algorithm is employed to detect abnormal subgraphs relative to the RG. By jointly analyzing the real and virtual graphs, the method accurately identifies the time points at which system anomalies occur. Using a typical aviation product as the case study, the proposed method was validated through Plant Simulation software. The results demonstrate that the method can effectively detect multiple types of system anomalies, providing new insights and innovative solutions for anomaly monitoring research in DMS, especially in combining real-time network topology changes with data-driven anomaly detection techniques.
{"title":"Integrating real and virtual graphs: a dual joint network method for anomaly detection in discrete manufacturing systems","authors":"Yi Li , Fan Zhang , Jingzheng Liu , Faping Zhang , Tianci Liu , Fanyue Zhou","doi":"10.1016/j.ress.2026.112246","DOIUrl":"10.1016/j.ress.2026.112246","url":null,"abstract":"<div><div>To address the challenges of anomaly monitoring arising from the increasing complexity and scale of Discrete Manufacturing Systems (DMS), this study proposes an anomaly monitoring method based on a Dual Joint Network (DJN), which integrates complex network theory with operational data. The model consists of two components: a real graph established through physical modeling and a virtual graph constructed through data-driven modeling. In the real graph, anomalies are detected by identifying abrupt changes in network topology relative to the Representative Graph (RG), which characterizes the normal operating state. In the virtual graph, an improved SpotLight algorithm is employed to detect abnormal subgraphs relative to the RG. By jointly analyzing the real and virtual graphs, the method accurately identifies the time points at which system anomalies occur. Using a typical aviation product as the case study, the proposed method was validated through Plant Simulation software. The results demonstrate that the method can effectively detect multiple types of system anomalies, providing new insights and innovative solutions for anomaly monitoring research in DMS, especially in combining real-time network topology changes with data-driven anomaly detection techniques.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"271 ","pages":"Article 112246"},"PeriodicalIF":11.0,"publicationDate":"2026-01-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146079477","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}
Pub Date : 2026-01-23DOI: 10.1016/j.ress.2026.112244
Yunfan Yang , Xuancheng Yuan , Ruichen Wang , Wai Kei Ao , Liang Ling , Paul Allen
Wear and rolling contact fatigue (RCF) severely deteriorate the tribology behaviour and material integrity of railway wheels, posing significant challenges to their health management. Hence, a deeper mechanistic understanding and the development of effective mitigation strategies are urgently required. In this study, a long-term locomotive wheel wear and RCF evolution prediction model were developed that incorporates the fully nonlinear dynamics of heavy-haul locomotive-track coupled system, the non-Hertzian wheel-rail frictional contact behaviour, and iterative updates of the evolving wear and RCF distributions. The numerical investigations indicated that wear and RCF growth of locomotive wheels are primarily caused by the prominent wheel/rail stresses during curving operations, and particularly aggravated at sharp curves. Subsequent numerical and field investigations verified two effective strategies for mitigating locomotive wheel wear and RCF development: (ⅰ) optimisation design of wheel profile using an innovative constrained multi-object optimisation (CMOO) method, and (ⅱ) enhancement of the Wheel Slide Protection (WSP) controller. The findings further suggested that these two countermeasures can substantially mitigate locomotive wheel wear and RCF progression by lowering wheel-rail tribological interaction and contact stress levels. Overall, this study provides valuable insight into the mechanisms governing wheel wear and RCF evolutions, and supports the enhancement of heavy-haul operational reliability through scientifically informed maintenance practices.
{"title":"Wear and rolling contact fatigue problems of locomotive wheels: Mechanisms and countermeasures","authors":"Yunfan Yang , Xuancheng Yuan , Ruichen Wang , Wai Kei Ao , Liang Ling , Paul Allen","doi":"10.1016/j.ress.2026.112244","DOIUrl":"10.1016/j.ress.2026.112244","url":null,"abstract":"<div><div>Wear and rolling contact fatigue (RCF) severely deteriorate the tribology behaviour and material integrity of railway wheels, posing significant challenges to their health management. Hence, a deeper mechanistic understanding and the development of effective mitigation strategies are urgently required. In this study, a long-term locomotive wheel wear and RCF evolution prediction model were developed that incorporates the fully nonlinear dynamics of heavy-haul locomotive-track coupled system, the non-Hertzian wheel-rail frictional contact behaviour, and iterative updates of the evolving wear and RCF distributions. The numerical investigations indicated that wear and RCF growth of locomotive wheels are primarily caused by the prominent wheel/rail stresses during curving operations, and particularly aggravated at sharp curves. Subsequent numerical and field investigations verified two effective strategies for mitigating locomotive wheel wear and RCF development: (ⅰ) optimisation design of wheel profile using an innovative constrained multi-object optimisation (CMOO) method, and (ⅱ) enhancement of the Wheel Slide Protection (WSP) controller. The findings further suggested that these two countermeasures can substantially mitigate locomotive wheel wear and RCF progression by lowering wheel-rail tribological interaction and contact stress levels. Overall, this study provides valuable insight into the mechanisms governing wheel wear and RCF evolutions, and supports the enhancement of heavy-haul operational reliability through scientifically informed maintenance practices.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"273 ","pages":"Article 112244"},"PeriodicalIF":11.0,"publicationDate":"2026-01-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146098487","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}
Pub Date : 2026-01-23DOI: 10.1016/j.ress.2026.112279
Feiyu Pang , Xiaokai Niu , Jie Su , Chengping Zhang
The safety assessment of operational shield tunnels involves complexity, randomness, and uncertainty. Traditional constant weight methods fail to account for the dynamic changes in structural state caused by the interaction among different defects. Therefore, this study proposes a novel safety evaluation framework for operational tunnels by integrating game theory and variable weight theory. This evaluation model is applied to four tunnel sections of Beijing Metro Line 8 and is compared with three other evaluation models (including conventional weighting methods, fuzzy comprehensive evaluation, etc.). Sensitivity analysis identified U31, U32, U33, and U52 as the key indicators affecting the tunnel’s structural safety. On-site investigation results demonstrate that the proposed model provides more accurate evaluations, thereby verifying its feasibility. Furthermore, this study has also established an evaluation framework that is applicable to the comprehensive assessment of the entire tunnel section. Evaluating an entire tunnel section as a single unit may conceal local high-risk areas, leading to inaccurate assessment results. It facilitated managers taking safeguard measures in a timely manner based on the evaluation results and ensuring the safety and reliability of the tunnel structure.
{"title":"Safety evaluation of operational metro shield tunnels using improved game theory and dynamic variable weight theory","authors":"Feiyu Pang , Xiaokai Niu , Jie Su , Chengping Zhang","doi":"10.1016/j.ress.2026.112279","DOIUrl":"10.1016/j.ress.2026.112279","url":null,"abstract":"<div><div>The safety assessment of operational shield tunnels involves complexity, randomness, and uncertainty. Traditional constant weight methods fail to account for the dynamic changes in structural state caused by the interaction among different defects. Therefore, this study proposes a novel safety evaluation framework for operational tunnels by integrating game theory and variable weight theory. This evaluation model is applied to four tunnel sections of Beijing Metro Line 8 and is compared with three other evaluation models (including conventional weighting methods, fuzzy comprehensive evaluation, etc.). Sensitivity analysis identified <em>U</em><sub>31</sub>, <em>U</em><sub>32</sub>, <em>U</em><sub>33</sub>, and <em>U</em><sub>52</sub> as the key indicators affecting the tunnel’s structural safety. On-site investigation results demonstrate that the proposed model provides more accurate evaluations, thereby verifying its feasibility. Furthermore, this study has also established an evaluation framework that is applicable to the comprehensive assessment of the entire tunnel section. Evaluating an entire tunnel section as a single unit may conceal local high-risk areas, leading to inaccurate assessment results. It facilitated managers taking safeguard measures in a timely manner based on the evaluation results and ensuring the safety and reliability of the tunnel structure.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"271 ","pages":"Article 112279"},"PeriodicalIF":11.0,"publicationDate":"2026-01-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146079552","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}
Pub Date : 2026-01-23DOI: 10.1016/j.ress.2026.112277
Xirui Li , Fairuz Izzuddin Romli , Syaril Azrad Md Ali , Amzari Zhahir , Junqi Tang
Aviation risk analysis can be a useful empirical foundation using narrative incident reports gathered by the Aviation Safety Reporting System (ASRS), but due to its long-form format, class imbalance, and domain-specific semantics, automated modelling can be a challenging problem. To respond to these challenges, this study develops a domain-adapted deep learning model built upon the Robustly Optimized Bidirectional Encoder Representations from Transformers pretraining approach (RoBERTa) for multi-label identification of contributing factors in aviation safety reports. The proposed model improves multi-label classification performance by integrating four modules: instruction-based large language models (LLMs) data augmentation to reduce imbalance, a merging module to jointly model the narrative text and metadata, a composite loss to strengthen robustness in case of label imbalance, and domain adaptive pretraining on corpora. The experimental results indicate that the model achieves reliable improvements, while ablation experiments further clarify impact of each module. Based on the predicted contributing factors, an N-K model is constructed to quantify interaction strength, and a Bayesian network is used to model directed risk propagation. By accounting for both structural coupling and propagation probability, the framework identifies and ranks risk pathways that correspond to plausible accident developments. A case study demonstrates that the proposed approach can extract high-order, multi-domain propagation paths from narrative data, enabling structured interpretation of plausible accident evolution patterns. Taken together, the proposed framework provides a pipeline that converts incident narratives into actionable safety information, offering a scalable and structured basis for proactive aviation risk analysis.
{"title":"A deep learning framework for aviation risk classification and high-order coupled risk modeling","authors":"Xirui Li , Fairuz Izzuddin Romli , Syaril Azrad Md Ali , Amzari Zhahir , Junqi Tang","doi":"10.1016/j.ress.2026.112277","DOIUrl":"10.1016/j.ress.2026.112277","url":null,"abstract":"<div><div>Aviation risk analysis can be a useful empirical foundation using narrative incident reports gathered by the Aviation Safety Reporting System (ASRS), but due to its long-form format, class imbalance, and domain-specific semantics, automated modelling can be a challenging problem. To respond to these challenges, this study develops a domain-adapted deep learning model built upon the Robustly Optimized Bidirectional Encoder Representations from Transformers pretraining approach (RoBERTa) for multi-label identification of contributing factors in aviation safety reports. The proposed model improves multi-label classification performance by integrating four modules: instruction-based large language models (LLMs) data augmentation to reduce imbalance, a merging module to jointly model the narrative text and metadata, a composite loss to strengthen robustness in case of label imbalance, and domain adaptive pretraining on corpora. The experimental results indicate that the model achieves reliable improvements, while ablation experiments further clarify impact of each module. Based on the predicted contributing factors, an N-K model is constructed to quantify interaction strength, and a Bayesian network is used to model directed risk propagation. By accounting for both structural coupling and propagation probability, the framework identifies and ranks risk pathways that correspond to plausible accident developments. A case study demonstrates that the proposed approach can extract high-order, multi-domain propagation paths from narrative data, enabling structured interpretation of plausible accident evolution patterns. Taken together, the proposed framework provides a pipeline that converts incident narratives into actionable safety information, offering a scalable and structured basis for proactive aviation risk analysis.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"271 ","pages":"Article 112277"},"PeriodicalIF":11.0,"publicationDate":"2026-01-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146079551","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}
Pub Date : 2026-01-22DOI: 10.1016/j.ress.2026.112268
Ali Shehadeh, Musab Abuaddous, Hamsa F. Nimer
We develop a multi-region SCGE–NEG model that embeds a capacity-constrained construction sector and time-critical logistics into a transport economy with endogenous migration and agglomeration externalities. Transport performance enters delivery prices via iceberg costs derived from network speeds and stochastic travel-time reliability. Construction production uses labor, equipment, and intermediate inputs (cement, aggregates, steel, bitumen) with queue-based delivery windows; late deliveries incur reliability penalties. We fuse probe-based travel-time distributions with multi-year e-ticketing records for hot-mix asphalt and aggregates, applying robust preprocessing to handle outliers and sensor noise (spec-based filters, trimming and winsorization of abnormal temperatures and weights, and cross-checks against project logs). A short–medium–long run solution cycle (goods, equilibrium, migration, and capital/entry) evaluates two policy families: (1) corridor upgrades & phasing, and (2) work-zone traffic management during project execution. Using a prefecture-scale testbed (47 regions; multi-sector IO base) and empirically plausible elasticities, pilot simulations indicate: on-time material delivery +14–22 percentage points, logistics cost −10–18%, contractor price inflation −4–7%, city-region GDP +1.1–2.6%, and net in-migration to upgraded hubs +1.2–2.9% over 10 years; unmanaged work-zone congestion raises project durations +6–11% and wage drift +3–5% in tight labor markets. Compared with speed-only models, adding reliability cuts late-penalty exposure −25–40% and improves welfare gains +0.2–0.5 pp. The framework produces sequencing recommendations (which link first, when) and procurement guidance (lane-closure policies, night work, staging) that jointly maximize welfare and project NPV under labor and supply-chain constraints.
{"title":"Dynamic SCGE–NEG for construction reliability: Probabilistic decision support for transport upgrades, work-zone operations, and regional labor mobility","authors":"Ali Shehadeh, Musab Abuaddous, Hamsa F. Nimer","doi":"10.1016/j.ress.2026.112268","DOIUrl":"10.1016/j.ress.2026.112268","url":null,"abstract":"<div><div>We develop a multi-region SCGE–NEG model that embeds a capacity-constrained construction sector and time-critical logistics into a transport economy with endogenous migration and agglomeration externalities. Transport performance enters delivery prices via iceberg costs derived from network speeds and stochastic travel-time reliability. Construction production uses labor, equipment, and intermediate inputs (cement, aggregates, steel, bitumen) with queue-based delivery windows; late deliveries incur reliability penalties. We fuse probe-based travel-time distributions with multi-year e-ticketing records for hot-mix asphalt and aggregates, applying robust preprocessing to handle outliers and sensor noise (spec-based filters, trimming and winsorization of abnormal temperatures and weights, and cross-checks against project logs). A short–medium–long run solution cycle (goods, equilibrium, migration, and capital/entry) evaluates two policy families: (1) corridor upgrades & phasing, and (2) work-zone traffic management during project execution. Using a prefecture-scale testbed (47 regions; multi-sector IO base) and empirically plausible elasticities, pilot simulations indicate: on-time material delivery +14–22 percentage points, logistics cost −10–18%, contractor price inflation −4–7%, city-region GDP +1.1–2.6%, and net in-migration to upgraded hubs +1.2–2.9% over 10 years; unmanaged work-zone congestion raises project durations +6–11% and wage drift +3–5% in tight labor markets. Compared with speed-only models, adding reliability cuts late-penalty exposure −25–40% and improves welfare gains +0.2–0.5 pp. The framework produces sequencing recommendations (which link first, when) and procurement guidance (lane-closure policies, night work, staging) that jointly maximize welfare and project NPV under labor and supply-chain constraints.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"271 ","pages":"Article 112268"},"PeriodicalIF":11.0,"publicationDate":"2026-01-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146079555","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}
Pub Date : 2026-01-22DOI: 10.1016/j.ress.2026.112269
Quan Qian , Jianghong Zhou , Bingchang Hou , Jie Wang , Hanmin Sheng , Jiusi Zhang
Numerous remaining useful life transfer prediction methods have been proposed to handle the issues of domain shift and knowledge transfer. However, the effectiveness of almost all these methods relies on the assumption that the sample dimensions of the source and target domains are equal. In practice, owing to differences in operating speeds and fault types, such a consistency assumption inevitably creates degradation information asymmetry between the two domains, thereby resulting in distorted measurement of intrinsic cross-domain data distribution. To bridge this gap, this study develops a new feature distribution adaptation method named dimension-mismatched adversarial network (DMAN) to offer a new modeling paradigm. In DMAN, a dimension selection rule based on the Nyquist sampling theorem and frequency resolution is established, enabling the distribution alignment to concentrate on genuine data bias caused by variations in operating conditions. An adaptive empirical mutual information calculator is designed to accurately assess the similarity of data distribution for both domains. On this basis, an adversarial training mechanism is proposed to learn domain-invariant intrinsic degradation features and achieve domain confusion. Experimental results on XJTU-SY and IEEE PHM 2012 Challenge datasets demonstrate the superiority of DMAN over several state-of-the-art approaches.
{"title":"Dimension-mismatched adversarial network: a new feature distribution adaptation method for rolling bearing RUL prediction","authors":"Quan Qian , Jianghong Zhou , Bingchang Hou , Jie Wang , Hanmin Sheng , Jiusi Zhang","doi":"10.1016/j.ress.2026.112269","DOIUrl":"10.1016/j.ress.2026.112269","url":null,"abstract":"<div><div>Numerous remaining useful life transfer prediction methods have been proposed to handle the issues of domain shift and knowledge transfer. However, the effectiveness of almost all these methods relies on the assumption that the sample dimensions of the source and target domains are equal. In practice, owing to differences in operating speeds and fault types, such a consistency assumption inevitably creates degradation information asymmetry between the two domains, thereby resulting in distorted measurement of intrinsic cross-domain data distribution. To bridge this gap, this study develops a new feature distribution adaptation method named dimension-mismatched adversarial network (DMAN) to offer a new modeling paradigm. In DMAN, a dimension selection rule based on the Nyquist sampling theorem and frequency resolution is established, enabling the distribution alignment to concentrate on genuine data bias caused by variations in operating conditions. An adaptive empirical mutual information calculator is designed to accurately assess the similarity of data distribution for both domains. On this basis, an adversarial training mechanism is proposed to learn domain-invariant intrinsic degradation features and achieve domain confusion. Experimental results on XJTU-SY and IEEE PHM 2012 Challenge datasets demonstrate the superiority of DMAN over several state-of-the-art approaches.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"271 ","pages":"Article 112269"},"PeriodicalIF":11.0,"publicationDate":"2026-01-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146079646","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}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号