Effect of Database Generation on Damage Consequences’ Assessment Based on Random Forests

Braidotti, Luca; Prpić-Oršić, Jasna; Valčić, Marko

doi:10.3390/jmse9111303

Cited by 4 publications

(7 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Other studies have introduced hybrid or multi-level real-time flooding risk evaluation [124]. Furthermore, recent studies have shown the potential for the application of data-driven techniques to the problem based on pre-calculated progressive flooding simulations [125][126][127][128][129].…”

Section: Predictionmentioning

confidence: 99%

See 1 more Smart Citation

A Review of Methods for Modelling Flooding, Its Progression and Outcome in Damaged Ships

Rodrigues

2024

JMSE

View full text Add to dashboard Cite

The timely and precise prediction of flooding progression and its eventual outcome in ships with breached hulls can lead to dramatic improvements in maritime safety through improved guidance for both emergency response and ship design. The traditional approach to assessing damage-induced flooding in both these stages, which also fully complies with statutory rules, is through static calculations. On the other hand, the application of models that simulate the flooding progression and the behaviour of flooded ships from, or close to, first principles allows for increased accuracy of the modelling of the phenomenon. This increase in accuracy can then be used to support advanced design for safety procedures. Furthermore, it can considerably enhance a ship’s capability for damage identification and inference-based logic for emergency decision support systems and marine accident response in general. This paper conducts a review of selected state-of-the-art methods, procedures, and case studies in recent years which aimed to model progressive flooding and damage ship behaviour and provide some explanations of fundamentals. Applications related to damage identification, the prediction of outcome/situation awareness, and flooding emergency response are also briefly discussed. The paper concludes with a brief reflection on salient gaps in the context of accelerating the development of these methods and their applicability.

show abstract

Section: Predictionmentioning

confidence: 99%

“…Likewise, Braidotti et al [128,129] have developed promising machine learning-based flooding progression methodologies.…”

Section: Identification Prediction and Responsementioning

confidence: 99%

A Review of Methods for Modelling Flooding, Its Progression and Outcome in Damaged Ships

Rodrigues

2024

JMSE

View full text Add to dashboard Cite

show abstract

“…The paper written by Braidotti et al [1] is related to the unsinkability problem; however, their scope is far beyond the classic single-case formulations. First, a scenario of the schematised accidental lateral ramming was considered and used for the generation of a damages database.…”

Section: Contributions' Overviewmentioning

confidence: 99%

Ship Dynamics and Hydrodynamics

Sutulo

Soares

2023

JMSE

View full text Add to dashboard Cite

show abstract

“…A progressive flood simulation was employed using a database-trained machine learning algorithm to assess the main consequences of a damage scenario. Furthermore, the application of random forests was proved to be suitable for assessing the final state of ships and damaged compartment settings and estimating flooding time [30].…”

Section: Introductionmentioning

confidence: 99%

Ship Flooding Time Prediction Based on Composite Neural Network

Yang

Yin

2023

JMSE

View full text Add to dashboard Cite

When a ship sailing on the sea encounters flooding events, quickly predicting the flooding time of the compartments in the damaged area is beneficial to making evacuation decisions and reducing losses. At present, decision-makers obtain flooding data through various sensors arranged on board to predict the time of compartment flooding. These data help with the calculation of the flooding time in emergency situations. This paper proposes a new approach to obtaining the compartment flooding time. Specifically in damage scenarios, based on Convolutional Neural Network and Recurrent Neural Network (CNN-RNN), using a composite neural network framework estimates the time when the compartment’s flooding water reaches the target height. The input of the neural network is the flooding images of the damaged compartment. Transfer learning is utilized in the paper. The ResNet18 model in Pytorch is used to extract the spatial information from the flooding images. The Long Short-Term Memory (LSTM) model is then applied to predict when the compartment flooding water reaches the target height. Experimental results show that, for the damaged compartment, the flooding time predicted by the neural network is 85% accurate while the others’ accuracy is more than 91%. Intuitively, when it comes to the actual flooding event, the composite neural network’s average prediction error for compartment flooding time is approximately 1 min. To summarize, these results suggest that the composite neural network proposed above can provide flooding information to assist decision-makers in emergency situations.

show abstract

Effect of Database Generation on Damage Consequences’ Assessment Based on Random Forests

Cited by 4 publications

References 19 publications

A Review of Methods for Modelling Flooding, Its Progression and Outcome in Damaged Ships

A Review of Methods for Modelling Flooding, Its Progression and Outcome in Damaged Ships

Ship Dynamics and Hydrodynamics

Ship Flooding Time Prediction Based on Composite Neural Network

Contact Info

Product

Resources

About