Reliability Engineering & System Safety最新文献

Developing a deep reinforcement learning model for safety risk prediction at subway construction sites

IF 9.4 1区工程技术

Reliability Engineering & System Safety Pub Date : 2025-02-04 DOI: 10.1016/j.ress.2025.110885

Zhipeng Zhou , Wen Zhuo , Jianqiang Cui , Haiying Luan , Yudi Chen , Dong Lin

{"title":"Developing a deep reinforcement learning model for safety risk prediction at subway construction sites","authors":"Zhipeng Zhou , Wen Zhuo , Jianqiang Cui , Haiying Luan , Yudi Chen , Dong Lin","doi":"10.1016/j.ress.2025.110885","DOIUrl":"10.1016/j.ress.2025.110885","url":null,"abstract":"<div><div>Underground construction work is heavily affected by surrounding hydrogeology, adjacent pipelines, and existing subway lines, which can lead to a high degree of uncertainty and generate safety risk on site. In order to overcome rigid thinking of causal factors within a structured framework and incorporate features of different accidents, this study adopted grounded theory for the investigation on factors contributing to workplace accidents in subway construction. The deep reinforcement learning model of double deep Q-network (DDQN) was developed for predicting subway construction safety risk, which integrated the advantage of reinforcement learning in decision making with the advantage of deep learning in objection perception. The findings denoted that DDQN performed better than other machine learning models inclusive of random forest, extreme gradient boosting, k-nearest neighbor, and support vector machine. Contributing factors relevant to subway construction accidents were quantitatively analyzed using permutation importance of attributes. It was beneficial for determining how the 37 contributing factors had negative effects on subway construction safety risk. Safety measures for risk reduction and controlling could be optimized according to permutation importance of individual contributing factor, which paved a new way for the promotion of safety management performance at subway construction sites.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"257 ","pages":"Article 110885"},"PeriodicalIF":9.4,"publicationDate":"2025-02-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143360215","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

Multi-perception graph convolutional tree-embedded network for aero-engine bearing health monitoring with unbalanced data

IF 9.4 1区工程技术

Reliability Engineering & System Safety Pub Date : 2025-02-03 DOI: 10.1016/j.ress.2025.110888

Dezun Zhao , Wenbin Cai , Lingli Cui

{"title":"Multi-perception graph convolutional tree-embedded network for aero-engine bearing health monitoring with unbalanced data","authors":"Dezun Zhao , Wenbin Cai , Lingli Cui","doi":"10.1016/j.ress.2025.110888","DOIUrl":"10.1016/j.ress.2025.110888","url":null,"abstract":"<div><div>In engineering, severely unbalanced data from aero-engine bearings leads data-driven methods to favor normal samples and disorganize decision boundaries, triggering poor performance. Although graph networks alleviate negative impact of unbalanced samples, they have limitations on single information transmission and graph adaptive updating. As such, a multi-perception graph convolutional tree-embedded network (MPGCTN) is developed. First, a dual-channel feature graph construction method is designed to convert high-dimensional mappings into feature distance and feature dynamic graphs, boosting diverse fault information. Then, multi-scale Chebyshev graph convolutional layers with multi-perception learning are constructed as the backbone network, capturing special and shared information through discrepancy and similarity constraints. Furthermore, a tree embedded decision layer is proposed as the rebuilt output layer to gradually recognize fault locations and sizes. Finally, a triple-loss training strategy is developed to update the parameters of the MPGCTN for deep feature extraction and hierarchical decision. Experimental results of two aero-engine bearing datasets demonstrate that the MPGCTN attains the classification accuracy of 97.54 % and 98.04 % with an unbalanced ratio of 20:1, outperforming state-of-the-art methods. From the above results, the MPGCTN exhibits excellent accuracy in gradually determining fault types and severities of aero-engine bearings with unbalanced data, consistent with the fundamental principles of maintenance.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"257 ","pages":"Article 110888"},"PeriodicalIF":9.4,"publicationDate":"2025-02-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143351041","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

Efficient value of information analysis for optimal monitoring placement of reinforced slopes by collaborative reliability updating

IF 9.4 1区工程技术

Reliability Engineering & System Safety Pub Date : 2025-02-01 DOI: 10.1016/j.ress.2025.110877

Hua-Ming Tian , Zi-Jun Cao , Dian-Qing Li , Yu Wang

{"title":"Efficient value of information analysis for optimal monitoring placement of reinforced slopes by collaborative reliability updating","authors":"Hua-Ming Tian , Zi-Jun Cao , Dian-Qing Li , Yu Wang","doi":"10.1016/j.ress.2025.110877","DOIUrl":"10.1016/j.ress.2025.110877","url":null,"abstract":"<div><div>Determining optimal monitoring placement (e.g., number and locations of monitoring variables) of engineering structures (e.g., reinforced slopes) is essential for field instrumentation and subsequent reliability assessments. Value of information (VoI) provides a decision-theoretic metric for effective structure monitoring planning by quantifying potential benefits of various monitoring schemes (e.g., possible combinations of monitoring candidates). However, computing VoI often involves a significant number of reliability analyses corresponding to possible monitoring outcomes (PMOs) from different monitoring schemes, which is computationally challenging. The challenge becomes more profound when reliability updating is conducted using a physics-based model (e.g., finite element model) and the failure probability is rare. This study proposes a collaborative reliability updating approach for VoI-based optimal monitoring placement of reinforced slopes. The proposed approach first decomposes the reliability updating problem into three reliability analysis problems, and then generates candidate sample pools that will be employed, collaboratively, for repeated estimations of the three decomposed reliability problems considering PMOs from different monitoring schemes. Using the proposed approach, a huge number (e.g., over a million) of reliability analyses can be accomplished in a cost-effective way. A simple numerical example and a geotechnical reinforced slope are adopted to illustrate the proposed approach for optimizing monitoring configurations considering different monitoring schemes.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"257 ","pages":"Article 110877"},"PeriodicalIF":9.4,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143287504","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

Corrigendum to “A closed-form continuous-depth neural-based hybrid difference features re-representation network for RUL prediction” [Reliability Engineering & System Safety 253C (2024) 110540]

IF 9.4 1区工程技术

Reliability Engineering & System Safety Pub Date : 2025-02-01 DOI: 10.1016/j.ress.2024.110635

Xuanlin Li , Yawei Hu , Hang Wang , Yongbin Liu , Xianzeng Liu , Huitian Lu

引用次数: 0

Wind hazard reliability assessment of a transmission tower-line system incorporating progressive collapse

IF 9.4 1区工程技术

Reliability Engineering & System Safety Pub Date : 2025-02-01 DOI: 10.1016/j.ress.2025.110878

Wen-Long Du , Xing Fu , Shuai Shao , Gang Li , Hong-Nan Li , Feng-Li Yang

{"title":"Wind hazard reliability assessment of a transmission tower-line system incorporating progressive collapse","authors":"Wen-Long Du , Xing Fu , Shuai Shao , Gang Li , Hong-Nan Li , Feng-Li Yang","doi":"10.1016/j.ress.2025.110878","DOIUrl":"10.1016/j.ress.2025.110878","url":null,"abstract":"<div><div>Wind-induced progressive collapse is the primary factor triggering large-scale failure of transmission tower line systems (TTLS), which seriously affects the reliable operation of the power system. The core innovation of this paper is incorporating the progressive collapse into the wind hazard reliability assessment of TTLS. First, a semi-analytical solution (SAS) is derived to quantify the nonlinear tensions in a multi-span conductor-insulator system, taking into account the high nonlinearity of insulators. During this process, a multi-dimensional nonlinear system of equations is constructed, with conductor reaction forces and insulator swinging displacements as variables. Subsequently, an efficient SAS-based progressive collapse analysis method is developed by simplifying the failed tower as a multi-segment rigid body model and coupling the two-dimensional overturning angles into the SAS, where the impact of the post-failure equilibrium on progressive collapse is highlighted. Afterwards, uncertain TTLS models are established, and the progressive collapse fragility is estimated using Monte Carlo simulation and SAS. A comprehensive sensitivity analysis is performed to rank the importance of uncertainty parameters affecting the model outcomes. Finally, both the yearly failure probability and reliability index before and after considering the progressive collapse are calculated. Numerical validation demonstrates the excellent reliability of the proposed method; neglecting progressive collapse leads to an overestimation of the reliability index.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"257 ","pages":"Article 110878"},"PeriodicalIF":9.4,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143353094","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

Review on modeling the societal impact of infrastructure disruptions due to disasters

IF 9.4 1区工程技术

Reliability Engineering & System Safety Pub Date : 2025-02-01 DOI: 10.1016/j.ress.2025.110879

Yongsheng Yang , Huan Liu , Ali Mostafavi , Hirokazu Tatano

{"title":"Review on modeling the societal impact of infrastructure disruptions due to disasters","authors":"Yongsheng Yang , Huan Liu , Ali Mostafavi , Hirokazu Tatano","doi":"10.1016/j.ress.2025.110879","DOIUrl":"10.1016/j.ress.2025.110879","url":null,"abstract":"<div><div>Infrastructure systems play a critical role in providing essential products and services for the functioning of modern society; however, they are vulnerable to disasters, and their service disruptions can cause severe societal impacts. To protect infrastructure from disasters and reduce potential impacts, great achievements have been made in modeling interdependent infrastructure systems in past decades. In recent years, scholars have gradually shifted their research focus to understanding and modeling societal impacts of disruptions considering the fact that infrastructure systems are critical because of their role in societal functioning, especially in situations of modern societies. Exploring how infrastructure disruptions impair society has become a key field of study. By comprehensively reviewing relevant studies, this paper demonstrated the definition and types of societal impact of infrastructure disruptions, and summarized the modeling approaches into four types: extended infrastructure modeling approaches, empirical approaches, agent-based approaches, and big data-driven approaches. For each approach, this paper organized relevant literature in terms of modeling ideas, advantages, and disadvantages. Furthermore, the four approaches were compared according to several criteria, including the input data, applicable societal impact types, spatial scales, and application contexts. Finally, this paper illustrated the challenges and future research directions in the field.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"257 ","pages":"Article 110879"},"PeriodicalIF":9.4,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143351040","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Resilience enhancement of cyber–physical distribution systems via mobile power sources and unmanned aerial vehicles

IF 9.4 1区工程技术

Reliability Engineering & System Safety Pub Date : 2025-02-01 DOI: 10.1016/j.ress.2024.110603

Meng Tian , Ziyang Zhu , Zhengcheng Dong , Le Zhao , Hongtai Yao

{"title":"Resilience enhancement of cyber–physical distribution systems via mobile power sources and unmanned aerial vehicles","authors":"Meng Tian , Ziyang Zhu , Zhengcheng Dong , Le Zhao , Hongtai Yao","doi":"10.1016/j.ress.2024.110603","DOIUrl":"10.1016/j.ress.2024.110603","url":null,"abstract":"<div><div>Mobile power sources (MPS) and unmanned aerial vehicles (UAV) are critical and flexible resources to enhance the resilience of cyber–physical distribution systems. However, they are usually independently planned and dispatched. In this paper, considering the cyber–physical interdependence, topology reconfiguration and planned distribution generator islanding, an allocation and dispatch strategy of MPSs and UAVs is proposed. Before an event, a two-stage stochastic optimization based allocation model is built to pre-position MPSs and UAVs considering the uncertainty of events. After the event, a dispatch model is proposed to identify routings of MPSs and UAVs to restore electricity services. Note that both the models are mixed-integer nonlinear three-dimensional (3D) problems. As the optimal service radius and height of a UAV are independent with other variables, these two models are decomposed into two parts, i.e., one part to calculate the optimal service radius and height, and the other to identify resilience enhancement strategy. Then the two models are transformed into a mixed-integer convex programming solved by Progressive Hedging algorithm and a mixed-integer second-order cone programming, respectively. The effectiveness is verified on the modified IEEE 33-node and 123-node test systems. Numerical results highlight the necessity of co-optimizing MPSs and UAVs on cyber–physical distribution systems.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"254 ","pages":"Article 110603"},"PeriodicalIF":9.4,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143173534","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

Investigation of ship collision accident risk factors using BP-DEMATEL method based on HFACS-SCA

IF 9.4 1区工程技术

Reliability Engineering & System Safety Pub Date : 2025-01-31 DOI: 10.1016/j.ress.2025.110875

Mingyang Guo , Miao Chen , Lihao Yuan , Zhihui Zhang , Jia Lv , Zhiyong Cai

{"title":"Investigation of ship collision accident risk factors using BP-DEMATEL method based on HFACS-SCA","authors":"Mingyang Guo , Miao Chen , Lihao Yuan , Zhihui Zhang , Jia Lv , Zhiyong Cai","doi":"10.1016/j.ress.2025.110875","DOIUrl":"10.1016/j.ress.2025.110875","url":null,"abstract":"<div><div>This paper proposes an objective analysis method based on historical data to comprehensively analyze the causal relationships and importance ranking of various risk factors in ship collision accidents. Firstly, 343 ship collision accident reports were investigated, and a HFACS-SCA model for analyzing risk factors in ship collisions was proposed based on the HFACS. Subsequently, Apriori and entropy methods were employed to analyze the data and obtain the weights of each risk factor. Finally, the BP-DEMATEL method was introduced to establish a quantitative evaluation system, resulting in the analysis of risk factors in collision accidents and the proposal of preventive measures. The results indicate that organizational factors are the main causal factors, while unsafe acts are the main effect factors in the main layers of HFACS-SCA. Among all risk factors, \"Over speed\" and \"Safety management system defects are identified as significant causes of accidents. The method proposed in this paper does not require expert judgment and obtains more objective and realistic results through historical data analysis. The findings of this study could assist relevant researchers in formulating preventive measures for ship collisions and enhancing the level of ship safety management.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"257 ","pages":"Article 110875"},"PeriodicalIF":9.4,"publicationDate":"2025-01-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143353092","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

Supply reliability allocation of natural gas pipeline network system based on composite allocation method

IF 9.4 1区工程技术

Reliability Engineering & System Safety Pub Date : 2025-01-31 DOI: 10.1016/j.ress.2025.110876

Shuaishuai Tang , Lei Hou , Yueqi Liu , Kai Yang , Xingshen Sun , Mincong Wang , Xiaoyu Zhang , Lumeng Jiang

{"title":"Supply reliability allocation of natural gas pipeline network system based on composite allocation method","authors":"Shuaishuai Tang , Lei Hou , Yueqi Liu , Kai Yang , Xingshen Sun , Mincong Wang , Xiaoyu Zhang , Lumeng Jiang","doi":"10.1016/j.ress.2025.110876","DOIUrl":"10.1016/j.ress.2025.110876","url":null,"abstract":"<div><div>The challenge of supply reliability allocation of natural gas pipeline network system cannot be solved by classical reliability allocation methods due to the complex mapping relationship between system supply reliability and unit reliability. In this paper, a composite allocation method for supply reliability allocation is proposed. Firstly, an evaluation indicator of supply reliability is established. The influencing factor and allocation principle of supply reliability are discussed. Secondly, hydraulic simulation is employed to calculate the system gas deficiency under different operating statuses. The contribution of unit operating status to system gas supply capacity are quantified. Thirdly, the influencing factor is prioritized, and influencing factors are further refined to evaluate unit technical ability. The scoring criteria is established, the comprehensive factor model is amended. Finally, a model for supply reliability allocation is established. To demonstrate the feasibility and superiority of the methodology, an example of a pipeline network in China is presented. The unit reliability requirement under system target reliability is clarified and system weak units are identified. The results demonstrate that the proposed composite allocation method is highly applicable to natural gas pipeline network. The majority of weak units are located in high operating load region, which is consistent with engineering.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"257 ","pages":"Article 110876"},"PeriodicalIF":9.4,"publicationDate":"2025-01-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143353093","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

A systematic resilience assessment framework for multi-state systems based on physics-informed neural network

IF 9.4 1区工程技术

Reliability Engineering & System Safety Pub Date : 2025-01-30 DOI: 10.1016/j.ress.2025.110866

Yuxuan He , Enrico Zio , Zhaoming Yang , Qi Xiang , Lin Fan , Qian He , Shiliang Peng , Zongjie Zhang , Huai Su , Jinjun Zhang

{"title":"A systematic resilience assessment framework for multi-state systems based on physics-informed neural network","authors":"Yuxuan He , Enrico Zio , Zhaoming Yang , Qi Xiang , Lin Fan , Qian He , Shiliang Peng , Zongjie Zhang , Huai Su , Jinjun Zhang","doi":"10.1016/j.ress.2025.110866","DOIUrl":"10.1016/j.ress.2025.110866","url":null,"abstract":"<div><div>Resilience is crucial for systems to maintain functionality under disturbances, especially in critical applications. However, current methods for assessing resilience in multi-state systems (MSS), particularly those modeled with Markov Repairable Processes (MRP), often face high computational costs and inefficiencies in handling complex dynamics. To address these issues, this paper proposes a systematic framework for resilience assessment of MSS whose recovery process is described as a MRP, integrated with enhanced Physics-Informed Neural Networks (PINN). In the first step of the framework, the computation of resilience indices is performed, based on the MRP of the MSS and considering the system evolution through vulnerable and recovery phases. In the second step of the framework, the enhanced PINN is integrated into the MRP solution. A typical standby MSS structure is analyzed based on the proposed framework. By gradient calibration and momentum-driving training, the computational cost is shown to be reduced by 92.4 %, compared to the eigenvector method of solution. The approach is adaptable to other safety-critical systems, offering a robust tool for more effective resilience evaluation and system optimization.</div></div>","PeriodicalId":54500,"journal":{"name":"Reliability Engineering & System Safety","volume":"257 ","pages":"Article 110866"},"PeriodicalIF":9.4,"publicationDate":"2025-01-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143287507","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