Dynamic Accident Network Model for Predicting Marine Accidents in Narrow Waterways Under Variable Conditions: A Case Study of the Istanbul Strait DOI Creative Commons
Serdar Yıldız, Özkan Uğurlu, Xinjian Wang

et al.

Journal of Marine Science and Engineering, Journal Year: 2024, Volume and Issue: 12(12), P. 2305 - 2305

Published: Dec. 14, 2024

Accident analysis models are crucial tools for understanding and preventing accidents in the maritime industry. Despite advances ship technology regulatory frameworks, human factors remain a leading cause of marine accidents. The complexity behavior, influenced by social, technical, psychological aspects, makes accident challenging. Various methods used to analyze accidents, but no single approach is universally chosen use as most effective. Traditional often emphasize errors, technical failures, mechanical breakdowns. However, hybrid models, which combine different approaches, increasingly recognized providing more accurate predictions addressing multiple causal factors. In this study, dynamic model based on Human Factors Analysis Classification System (HFACS) Bayesian Networks proposed predict estimate risks narrow waterways. utilizes past data expert judgment assess potential ships encounter when navigating these confined areas. Uniquely, enables prediction probabilities under varying operational conditions, offering practical applications such real-time risk estimation vessels before entering Istanbul Strait. By insights, supports traffic operators implementing preventive measures enter high-risk zones. results study can serve decision-support system not only VTS operators, shipmasters, company representatives also national international stakeholders industry, aiding both probability development measures.

Language: Английский

Resilience-based complex system early design using dynamic Copula Bayesian network: Heave compensation hydraulic system design as a case study DOI
Chao Zhang, Yaohui Lu, Rentong Chen

et al.

Ocean Engineering, Journal Year: 2025, Volume and Issue: 320, P. 120314 - 120314

Published: Jan. 10, 2025

Language: Английский

Citations

7
Integration of MIMAH and Fuzzy Bayesian Networks for risk analysis in chemical tanker loading operations DOI
Cenk Ay

Journal of Marine Engineering & Technology, Journal Year: 2025, Volume and Issue: unknown, P. 1 - 19

Published: Feb. 13, 2025

This study provides a systematic risk assessment approach for chemical tanker loading operations, focusing on high-risk scenario identified through operational data from model vessel. To address the complexities of transportation, hybrid methodology combining Methodology Identification Major Accident Hazards (MIMAH) and Fuzzy Bayesian Network (FBN) analysis was developed. MIMAH's structured framework systematically identifies critical events using Bow-Tie (BT) diagram, integrating Fault Tree (FT) Event (ET) providing thorough breakdown potential accident pathways. BT structure converted into (BN) to improve probability estimations by incorporating conditional dependencies expert-driven fuzzy logic, particularly where historical limited. The further employed dual-method sensitivity analysis, Fussell-Vesely (FV) importance measures Improvement Index (II), identify improvement-prone basic (BEs). Key findings highlight dominance human error in events, manifold connection failures incorrect valve alongside mechanical vulnerabilities with significant improvement potential. extends ARAMIS principles maritime contexts, reliability-based fuzzy-based estimation methods detailed adaptable that enhances safety resilience hazardous transport.

Language: Английский

Citations

1
Reliability analysis of subsea pipeline system based on fuzzy polymorphic bayesian network DOI Creative Commons
Chao Liu,

Chuankun Zhou,

Hongyan Wang

et al.

Scientific Reports, Journal Year: 2025, Volume and Issue: 15(1)

Published: April 4, 2025

Subsea pipeline system faces significant challenges in practical engineering applications, including complexity, environmental variability, and limited historical data. These factors complicate the accurate estimation of component failure rates, leading to fault polymorphism inherent uncertainty. To address these challenges, this study proposes a reliability analysis method based on Fuzzy Polymorphic Bayesian Network (FPBN). The approach utilizes multi-state tree construct polymorphic (BN), integrating traditional BN techniques with consideration multiple states fuzzy rates. This extension allows network handle uncertainties such as imprecise data unclear logical relationships. is applied subsea risk by developing model. Through quantitative analysis, probability calculated. Reverse diagnosis then conducted determine posterior probabilities root nodes identify vulnerabilities. results demonstrate that FPBN effectively addresses ambiguity uncertainty providing robust framework applications.

Language: Английский

Citations

0
Causation Analysis of Marine Traffic Accidents Using Deep Learning Approaches: A Case Study from China’s Coasts DOI Creative Commons
Zelin Zhao,

Xingyu Liu,

Lin Feng

et al.

Systems, Journal Year: 2025, Volume and Issue: 13(4), P. 284 - 284

Published: April 12, 2025

In response to the increasing frequency of maritime traffic accidents along China’s coast, this study develops an accident-cause analysis framework that integrates optimized Bidirectional Encoder Representations from Transformers (BERT) with a Long Short-Term Memory network (BiLSTM), combined Apriori association rule algorithm. Systematic performance comparisons demonstrate BERT + BiLSTM architecture achieves superior unstructured-text-processing capability, attaining 89.8% accuracy in classification. The hybrid enables comprehensive investigation complex interactions among human factors, vessel characteristics, environmental conditions, and management practices through multidimensional accident reports. Our findings identify improper operations, fatigue-related issues, illegal modifications, inadequate as primary high-risk factors while revealing multi-factor interaction patterns significantly influence severity. Compared traditional single-factor methods, proposed shows marked improvements Natural Language Processing (NLP) efficiency, classification precision, systematic interpretation cross-factor correlations. This integrated approach provides authorities scientific evidence develop targeted prevention strategies optimize safety systems, thereby enhancing governance coastline.

Language: Английский

Citations

0
A multi‐perspective process safety risk assessment with hybrid risks DOI

Tahere Vafaee,

Mohammad Ali Saniee Monfared

Risk Analysis, Journal Year: 2025, Volume and Issue: unknown

Published: April 18, 2025

Abstract In this paper, we assert that the process safety risks vary based on identity of stakeholders involved, for example, employees, management, regulators, community members, insurance companies, and environment. These differ in perceptions, magnitudes, ramifications across an array stakeholders. Hence, risk assessment taken from a single perspective, as is often case, inadequate perhaps misleading. Instead, more realistic approach multi‐perspective by considering interactions existing among different perspectives building concurrent compatible models explicitly. This marks first innovation current research work. The second centers hybrid nature analysis. We recognize distinction between impacting human well‐being affecting facilities, properties, capital assets, introduces safety‐facility to address types risks. Still, developing multiple represent complex, time‐consuming, tedious, very costly. addition, results may become incompatible, confusing To avoid such difficulties, comprehensive model developed initially, which, while impractical itself, allows extraction practical perspective‐based through reduction. methodology was illustrated validated examining city gas pressure reduction station 12 perspectives, illustrating highlighting necessity accurate However, widely applicable areas, not limited gate (CGS). Furthermore, twelve considered are specific context CGS case suburb Tehran other situations. By incorporating these practices, organizations can ensure comprehensive, inclusive, risks, ultimately leading better management decision‐making.

Language: Английский

Citations

0
Dynamic risk analysis of tank cleaning operations using bow-tie-based fuzzy Bayesian network DOI

Taylan Güler,

Cenk Ay, İsmail Çiçek

et al.

Journal of Marine Engineering & Technology, Journal Year: 2024, Volume and Issue: 24(1), P. 22 - 39

Published: Aug. 27, 2024

This study aims to enhance maritime safety by focusing on risk analysis in tank cleaning operations, specifically targeting the of spillages. The Bow-Tie-based Fuzzy Bayesian Network (BT-FBN) methodology was employed conduct an equipment-based dynamic analysis. A real chemical tanker selected as model vessel ensure realistic calculations, and 'Loss Containment' (LoC) identified critical event for comprehensive BT utilised identify systematically map potential hazards. Considering temporal dynamics risk, a performed using four-month data from vessel, fuzzy logic applied address uncertainties data. findings indicated steady increase 'LoC' 4.33% 11.86% over four months, consistent with observed accident rates (8.57%). Furthermore, likelihood consequential events, such 'Delay Operations & Near Miss' (C-1), showed similar trend, ranging 3.69% 10.09%. Notably, underscored role 'Human Error Management Failure' (IE4) contributing most significantly 'LoC', 2.86% 8.77%, emphasising importance addressing human factors mitigate environmental pollution.

Language: Английский

Citations

3
Comprehensive analysis of ship sinking accidents using Bayesian network DOI
Dwitya Harits Waskito, Ludfi Pratiwi Bowo, Feronika Sekar Puriningsih

et al.

Australian Journal of Maritime & Ocean Affairs, Journal Year: 2024, Volume and Issue: unknown, P. 1 - 24

Published: Aug. 8, 2024

Sinking accidents can cause catastrophic losses in human, economic resources and environmental damage. are one of the most common maritime that occur Indonesia. Therefore, this study aims to analyze cause-and-effect probability ship sinkings Indonesian waters. Bayesian Network method accident factor level used causal factors sinking accidents. The results found Loss Stability vessel is sensitive node. An additional take from investigated half occurred on passenger vessels, ships over 25 years old prone due corrosion lack proper maintenance. While 70.5% caused by human error majorly triggered unsafe acts. real-case analysis also performed reveals eliminating Human Error ship's associated reduce 17.2%.

Language: Английский

Citations

2
Development of Spatial Clustering Method and Probabilistic Prediction Model for Maritime Accidents DOI Creative Commons
Cheol-Won Choe, Suhwan Lim,

Dong Jun Kim

et al.

Applied Ocean Research, Journal Year: 2024, Volume and Issue: 154, P. 104317 - 104317

Published: Nov. 16, 2024

Language: Английский

Citations

1
Navigation risk assessment for ocean-going ships in the north pacific ocean based on an improved dynamic Bayesian network model DOI
Yingying Wang, Longxia Qian, Mei Hong

et al.

Ocean Engineering, Journal Year: 2024, Volume and Issue: 315, P. 119804 - 119804

Published: Nov. 20, 2024

Language: Английский

Citations

1
Research on Response Strategies for Inland Waterway Vessel Traffic Risk Based on Cost-Effect Trade-Offs DOI Creative Commons
Yanyi Chen,

Ziyang Ye,

Tao Wang

et al.

Journal of Marine Science and Engineering, Journal Year: 2024, Volume and Issue: 12(9), P. 1659 - 1659

Published: Sept. 16, 2024

Compared to maritime vessel traffic accidents, there is a scarcity of available, and only incomplete, accident data for inland waterway accidents. Additionally, the characteristics different segments vary significantly, factors affecting navigation safety risks their mechanisms may also differ. Meanwhile, in recent years, extreme weather events have been frequent waterways, has clear trend towards larger vessels, bringing about new hazards management challenges. Currently, research on mainly focuses risk assessment, with scarce quantitative studies mitigation measures. This paper proposes method improving safety, based cost-effectiveness trade-off approach mitigate The links effectiveness cost measures constructs comprehensive cost-benefit evaluation model using fuzzy Bayesian quantification conversion techniques, considering reduction effects under uncertain conditions various costs they incur. Taking upper, middle, lower reaches Yangtze River as examples, this evaluates key provides most cost-effective strategies. Findings reveal that, even if waterways share same sources, due environmental differences. Moreover, no inherent correlation between best-performing terms benefits lowest-cost measures, nor are necessarily recommended. proposed case provide theoretical support scientifically formulating complex environments offer guidance departments determine future work directions.

Language: Английский

Citations

0