| 324 | 0 | 5 |
| Downloads | Citas | Reads |
This study presents a methodology for modeling ship operation safety risks using a data-driven complex network approach. The process begins with an analysis of Port State Control(PSC) defect data, utilizing the Apriori algorithm to mine frequent itemsets and identify association rules among various risk factors. This step quantifies the non-linear interactive effects between system components. The support, confidence, and lift metrics derived from these rules are then synthesized to create an interaction strength matrix. This matrix serves as the adjacency matrix for developing a Functional Resonance Analysis Method(FRAM) model, which represents the ship's operational safety system as a complex network, where nodes correspond to system components and weighted edges indicate their interaction strength. A Graph Convolutional Network(GCN) is subsequently applied to learn the latent features of this network structure. By processing both the node features and the adjacency matrix, the GCN identifies key nodes with high centrality and influence, effectively restructuring the interaction network to emphasize the most critical components. Subsequently, the Depth-First Search(DFS) algorithm is employed on the refined graph to systematically trace critical risk propagation paths, simulating how initial failures can resonate and cascade through the system. The influence of each component factor is calculated based on its position and connections within these identified paths. Simulation results derived from the PSC data confirm that a ship's unsafe state arises from the coupled interaction of both internal and external factors. The model successfully identifies several critical resonance paths where risks are significantly amplified. Notably, the interaction path between the firefighting system and the ship's structural integrity emerged as a critical resonance loop, indicating that a deficiency in one can severely compromise the other and escalate overall risk. The study demonstrates the model's sensitivity, as it dynamically generates distinct risk path dependencies when provided with varying data inputs. This validates its effectiveness in simulating risk emergence in systems characterized by uncertain structural interactions.
[1] CORADDU A,ONETO L,DE MAYA B N,et al.Determining the most influential human factors in maritime accidents:a data-driven approach[J].Ocean Engineering,2020,211:107588.
[2] 冯如，李帅.FDA事故致因模型在海上船舶碰撞事故安全管理中的运用研究[J].中国安全生产科学技术，2019,15(7):179-185.FENG R,LI S.Application of FDA accident causation model in safety management of marine ship collision accident[J].Journal of Safety Science and Technology,2019,15(7):179-185.
[3] CHEN J H,DI Z J,SHI J,et al.Marine oil spill pollution causes and governance:a case study of Sanchi tanker collision and explosion[J].Journal of Cleaner Production,2020,273(4):122978.
[4] MD HANAFIAH R,ZAINON N S,KARIM N H,et al.A new evaluation approach to control maritime transportation accidents:a study case at the Straits of Malacca[J].Case Studies on Transport Policy,2022,2:751-763.
[5] 张笛，万程鹏，严新平.基于事故特征分析的长江碍航风险研究[J].中国航海，2013,36(2):94-99.ZHANG D,WAN C P,YAN X P.Navigation hindering risk assessment of the yangtze river based on analysis of accident characteristics[J].Navigation of China,2013,36(2):94-99.
[6] 冯海商，陈厚忠.基于动态贝叶斯网络的客滚船航行风险预测[J].武汉理工大学学报，2023,45(9):96-102.FENG H S,CHEN H Z.Risk prediction of ro-ro passenger ships navigation based on dynamic Bayesian network[J].Journal of Wuhan University of Technology,2023,45(9):96-102.
[7] 邵波，刘巧，柯善钢，等.融合LDA-BN的船舶碰撞事故致因分析[J].安全与环境学报，2025,25(1):157-164.SHAO B,LIU Q,KE S G,et al.Cause analysis of ship collision accidents using LDA-BN methodology[J].Journal of Safety and Environment,2025,25(1):157-164.
[8] HOLLNAGEL E.Barriers and accident prevention[M].Abingdon,Oxon:Routledge,2016.
[9] 张晓全，吴贵锋.功能共振事故模型在可控飞行撞地事故分析中的应用[J].中国安全生产科学技术，2011,7(4):65-70.ZHANG X Q,WU G F.Application of FRAM model in CFIT accident analysis[J].Journal of Safety Science and Technology,2011,7(4):65-70.
[10] YU Y R,LV W F,HE Z P,et al.Framework for process risk analysis of maritime accidents based on resilience theory:a case study of grounding accidents in Arctic waters[J].Reliability Engineering & System Safety,2024,249:110202.
[11] 胡甚平，焦喜鑫，张欣欣，等.复杂性系统视角下船员不安全行为功能共振分析模型[J].安全与环境学报，2023,23(4):1177-1184.HU S P,JIAO X X,ZHANG X X,et al.A functional resonance analysis model of crew unsafe act in a complex systems perspective[J].Journal of Safety and Environment,2023,23(4):1177-1184.
[12] 张靖雯，马晓雪，刘阳，等.基于FRAM-FAHP法的船舶碰撞事故致因分析[J].安全与环境工程，2021,28(1):29-35.ZHANG J W,MA X X,LIU Y,et al.A causation analysis of ship collision accidents based on FRAMFAHP method[J].Safety and Environmental Engineering,2021,28(1):29-35.
[13] GUO Y L,JIN Y X,HU S P,et al.Risk evolution analysis of ship pilotage operation by an integrated model of FRAM and DBN[J].Reliability Engineering & System Safety,2023,229:108850.
[14] FU S S,YU Y R,CHEN J H,et al.Towards a probabilistic approach for risk analysis of nuclear-powered icebreakers using FMEA and FRAM[J].Ocean Engineering,2022,260:112041.
[15] 胡立伟，吕一帆，赵雪亭，等.基于数据驱动的交通事故伤害程度影响因素及其耦合关系研究[J].交通运输系统工程与信息，2022,22(5):117-124,134.HU L W,LÜ Y F,ZHAO X T,et al.Influence factors and coupling relationship of traffic accident injury degree based on a data-driven approach[J].Journal of Transportation Systems Engineering and Information Technology,2022,22(5):117-124,134.
[16] 魏凌峰，姜文龙.基于支持向量机的交通事故影响因素分析[J].山东交通科技，2022(1):84-87.WEI L F,JIANG W L.Analysis of influencing factors of traffic accident severity based on support vector machine[J].Shandong Transportation Science and Technology,2022(1):84-87.
[17] 黄常海，沈佳，朱冉超，等.基于C5.0决策树的船舶交通事故致因分析模型及应用[J].中国安全科学学报，2022,32(10):90-99.HUANG C H,SHEN J,ZHU R C,et al.Causation analysis model of marine traffic accidents based on C5.0 decision tree and its application[J].China Safety Science Journal,2022,32(10):90-99.
[18] 郑霞忠，欧妍希，田丹，等.集成学习框架下水电工程施工事故致因因果发现[J/OL].安全与环境学报：1-12[2025-06-17].https://doi.org/10.13637/j.issn.1009-6094.2024.1090.ZHENG X Z,OU Y X,TIAN D,et al.Causal discovery of causes of hydroelectric engineering construction accidents under an integrated learning framework[J/OL].Journal of Safety and Environment:1-12[2025-06-17].https://doi.org/10.13637/j.issn.1009-6094.2024.1090.
[19] JIA M S,GABRYS B,MUSIAL K.A network science perspective of graph convolutional networks:a survey[J].IEEE Access,2023,11:39083-39122.
[20] 李毅豪，洪征，林培鸿.基于深度优先搜索的模糊测试用例生成方法[J].计算机科学，2021,48(12):85-93.LI Y H,HONG Z,LIN P H.Fuzzing test case generation method based on depth-first search[J].Computer Science,2021,48(12):85-93.
[21] 张嘉易，尚文利，刘英.基于DFS的工程图多层次图块信息提取[J].计算机工程，2012,38(11):192-195.ZHANG J Y,SHANG W L,LIU Y.Multi-level segment information extraction in engineering drawings based on depth first search[J].Computer Engineering,2012,38(11):192-195.
[22] IRSHAD L,DEMIREL H O,TUMER I Y.Automated generation of fault scenarios to assess potential human errors and functional failures in early design stages[J].Journal of Computing and Information Science in Engineering,2022,20(5):1-31.
[23] KIPF T N,WELLING M.Semi-supervised classification with graph convolutional networks[PP/OL].(2017-02-22)[2025-06-17].https://arxiv.org/abs/1609.02907.
Basic Information:
DOI：10.13637/j.issn.1009-6094.2025.0663
China Classification Code:U698
Citation Information:
[1]HU Shenping,WANG Shengjun,XI Xiuting ,et al.Functional resonance model of ship operation safety risks based on complex networks[J].Journal of Safety and Environment,2026,26(03):823-833.DOI:10.13637/j.issn.1009-6094.2025.0663.
Fund Information:
国家重点研发计划项目(2021YFC2801005); 国家自然科学基金项目(52272353)