A framework for quantitative analysis of the causation of grounding accidents in arctic shipping

Reliability Engineering & System Safety, Journal Year: 2022, Volume and Issue: 226, P. 108706 - 108706

Published: July 1, 2022

Language: Английский

---

A machine learning method for the evaluation of ship grounding risk in real operational conditions

Reliability Engineering & System Safety, Journal Year: 2022, Volume and Issue: 226, P. 108697 - 108697

Published: July 2, 2022

Ship groundings may often lead to damages resulting in oil spills or ship flooding and subsequent capsizing. Risks can be estimated qualitatively through experts' judgment quantitatively the analysis of maritime traffic data. Yet, studies using big data remain limited. In this paper, we present a analytics method for evaluation grounding risk real environmental conditions. The makes use streams from Automatic Identification System (AIS), nowcast data, seafloor depth General Bathymetric Chart Oceans (GEBCO). evasive action Ro-Pax passenger ships operating shallow waters is idealized under various patterns that link side - forward scenarios. Consequently, an Avoidance Behaviour-based Grounding Detection Model (ABGD-M) introduced identify potential scenarios, probabilistic quantified at observation points along routes voyages. applied on over 2.5 years ice-free period Gulf Finland. Results indicate estimation extremely diverse depends voyage routes, points, operational It concluded proposed assist with (1) better identification critical scenarios are underestimated existing accident databases; (2) improved understanding avoidance behaviours conditions; (3) profile life cycle fleet operations (4) waterway complexity indices vulnerability.

Language: Английский

---

Ship trajectory prediction based on machine learning and deep learning: A systematic review and methods analysis

Engineering Applications of Artificial Intelligence, Journal Year: 2023, Volume and Issue: 126, P. 107062 - 107062

Published: Sept. 4, 2023

Ship trajectory prediction based on Automatic Identification System (AIS) data has attracted increasing interest as it helps prevent collision accidents and eliminate potential navigational conflicts. Therefore, is necessary urgent to conduct a systematic analysis of all the methods help reveal their advantages ensure safety at sea in different scenarios. It particularly important significant within context unmanned ships forming new hybrid maritime traffic together with manned future. This paper aims comparative up-to-date ship algorithms machine learning deep methods. To do so, five classical (i.e., Kalman Filter, Gaussian Process Regression, Support Vector Random Forest, Back Propagation Network) eight Recurrent Neural Networks, Long Short-Term Memory, Bi-directional Gate Unit, Sequence Sequence, Spatio-Temporal Graph Convolutional Network, Transformer) are thoroughly analysed compared from algorithm essence applications excavate features adaptability for ships. The findings characteristics various provide valuable implications stakeholders guide best-fit choice particular method solution under specific circumstance. also makes contributions extraction research difficulties corresponding solutions that put forward development future research.

Language: Английский

---

A deep learning method for the prediction of ship fuel consumption in real operational conditions

Engineering Applications of Artificial Intelligence, Journal Year: 2023, Volume and Issue: 130, P. 107425 - 107425

Published: Dec. 22, 2023

In recent years, the European Commission and International Maritime Organization (IMO) implemented various operational measures policies to reduce ship fuel consumption related emissions. The effectiveness of these relies upon developing accurate predictive models encompassing influence real conditions. This paper presents a deep learning method for prediction consumption. utilizes big data analytics from sensors, voyage reporting hydrometeorological data, comprising 266 variables made available following sea trials Kamsarmax bulk carrier Laskaridis Shipping Co. Ltd. A variable importance estimation model using Decision Tree (DT) is used understand underlying relationships in dataset. Consequently, developed sailing speed, heading, displacement/draft, trim, weather, conditions, etc. on (SFC). achieved by incorporating attention mechanism into Bi-directional Long Short-Term Memory (Bi-LSTM) network. potential new demonstrated training streams corresponding rates as well internal external comprehensive comparison with existing methods indicates that Bi-LSTM best fit when high frequency data. It concluded subject further testing validation could be development decision support systems monitoring environmentally sustainable operations.

Language: Английский

---

An object-oriented Bayesian network model for the quantitative risk assessment of navigational accidents in ice-covered Arctic waters

Reliability Engineering & System Safety, Journal Year: 2023, Volume and Issue: 238, P. 109459 - 109459

Published: June 19, 2023

Merchant ship operations in the ice-covered Arctic waters may encounter traditional navigational accident risks (i.e., grounding, collision, etc.) and from sea ice, such as besetting ice. However, describing, modeling, quantifying multiple ice navigation are challenges maritime risk assessment perspective. This paper proposes an object-oriented Bayesian network (OOBN) model for quantitative of accidents waters. The OOBN makes use database Lloyd's intelligence investigation reports. proposed decomposes into five levels based on causation theory: environment, unsafe condition, act, probability accident, consequence accident. Consequently, ship–ice collision selected cases to interpret factors identification, analysis, evaluation. results demonstrate that (1) is highest grounding accidents, followed by waters; (2) speed condition critical mutual these four scenarios; (3) influencing specific identified propose corresponding control options. can be used

Language: Английский

---

Maritime traffic partitioning: An adaptive semi-supervised spectral regularization approach for leveraging multi-graph evolutionary traffic interactions

Transportation Research Part C Emerging Technologies, Journal Year: 2024, Volume and Issue: 164, P. 104670 - 104670

Published: May 27, 2024

Maritime situational awareness (MSA) has long been a critical focus within the domain of maritime traffic surveillance and management. The increasing complexities ship traffic, originating from sophisticated multi-attribute interactions among multiple ships, coupled with continuous evolution dynamics, pose significant challenges in attaining accurate MSA, particularly complex port waters. This study is dedicated to establishing an advanced methodology for partitioning aimed at enhancing pattern interpretability strengthening anti-collision risk Specifically, three interaction measure metrics, including conflict criticality, spatial distance, approaching rate, are initially introduced quantify different aspects spatiotemporal ships. Subsequently, semi-supervised spectral regularization framework devised adeptly accommodate both information prior knowledge derived historic structures. facilitates segmentation regional into clusters, wherein ships same cluster exhibit high temporal stability, connectivity, compactness, convergent motion. Meanwhile, adaptive hyperparameter selection model engineered seek optimal outcomes across diverse scenarios, while also incorporating user preferences specific indicators. Comprehensive experiments using AIS data Ningbo-Zhoushan Port undertaken thoroughly assess models' efficacy. Research findings case analyses comparisons distinctly showcase capability proposed approach successfully deconstruct complexity, capture high-risk zones, strengthen strategic safety measures. Consequently, holds promise advancing intelligence systems facilitating automation

Language: Английский

---

Analysis of human errors in human-autonomy collaboration in autonomous ships operations through shore control experimental data

Reliability Engineering & System Safety, Journal Year: 2024, Volume and Issue: 246, P. 110080 - 110080

Published: March 14, 2024

Human-autonomy collaboration plays a pivotal role in the development of Maritime autonomous surface ships (MASS), as Shore control center (SCC) operators may engage loop by directly operating MASS, or, supervisory loop, monitoring MASS and taking over when needed. Thus, efficient human performance during takeover operation is crucial for safety operations. However, since still early phase development, mechanism errors unknown, data on human-autonomy collaborative scarce. Human reliability analysis (HRA) aims to assess qualitatively quantitatively, widely used various complex systems help analysis. This study dedicated incorporating advanced HRA methods elements identify quantify collision avoidance scenarios. It presents virtual experimental results, combined with theoretical error identification assessment methods. At first, we apply Human-System Interaction Autonomy (H-SIA) method potential errors; secondly, relevant Performance Shaping Factors (PSFs) including Experience, Boredom, Task complexity, Available time Pre-warning, measures errors, implement them experiment based full-scale ferry research vessel called milliAmpere2. Finally, build Bayesian Network (BN) present causal probabilistic relationships between PSFs through data. The results show that available has highest impact operators, followed task complexity pre-warning. Boredom does not significant sole unless time. Experience performance. In addition relevance safe operational design developed benefits other systems. BN model shows adaptability probabilities, practical significance integrating into existing methodologies

Language: Английский

---

Systems driven intelligent decision support methods for ship collision and grounding prevention: Present status, possible solutions, and challenges

Reliability Engineering & System Safety, Journal Year: 2024, Volume and Issue: unknown, P. 110489 - 110489

Published: Sept. 1, 2024

Language: Английский

---

Optimizing anti-collision strategy for MASS: A safe reinforcement learning approach to improve maritime traffic safety

Ocean & Coastal Management, Journal Year: 2024, Volume and Issue: 253, P. 107161 - 107161

Published: April 29, 2024

Language: Английский

---

Deep bi-directional information-empowered ship trajectory prediction for maritime autonomous surface ships

Transportation Research Part E Logistics and Transportation Review, Journal Year: 2023, Volume and Issue: 181, P. 103367 - 103367

Published: Dec. 6, 2023

It is critical to have accurate ship trajectory prediction for collision avoidance and intelligent traffic management of manned ships emerging Maritime Autonomous Surface Ships (MASS). Deep learning methods based on AIS data emerged as a contemporary maritime transportation research focus. However, concerns about its accuracy computational efficiency widely exist across both academic industrial sectors, necessitating the discovery new solutions. This paper aims develop approach called Bi-Directional Information-Empowered (DBDIE) by utilising integrated multiple networks an attention mechanism address above issues. The DBDIE model extracts valuable features fusing Bi-directional Long Short-Term Memory (Bi-LSTM) Gated Recurrent Unit (Bi-GRU) neural networks. Additionally, weights two bi-directional units are optimised using mechanism, final results obtained through weight self-adjustment mechanism. effectiveness proposed verified comprehensive comparisons with state-of-the-art deep methods, including Neural Network (RNN), (LSTM), (GRU), Bi-LSTM, Bi-GRU, Sequence (Seq2Seq), Transformer experimental demonstrate that achieves most satisfactory outcomes than all other classical providing solution improving predicting trajectories, which becomes increasingly important in era safe navigation mixed MASS. As result, findings can aid development implementation proactive preventive measures avoid collisions, enhance efficiency, ensure safety.

Language: Английский

---