Towards probabilistic models for the prediction of a ship performance in dynamic ice

Research output: Contribution to journalArticleScientificpeer-review

Standard

Towards probabilistic models for the prediction of a ship performance in dynamic ice. / Montewka, Jakub; Goerlandt, Floris; Kujala, Pentti; Lensu, Mikko.

In: Cold Regions Science and Technology, Vol. 112, 2015, p. 14-28.

Research output: Contribution to journalArticleScientificpeer-review

Harvard

APA

Vancouver

Author

Bibtex - Download

@article{7cea38f240c4448e8bbbb1f91f2c4496,
title = "Towards probabilistic models for the prediction of a ship performance in dynamic ice",
abstract = "For safe and efficient exploitation of ice-covered waters, knowledge about ship performance in ice is crucial. The literature describes numerical and semi-empirical models that characterize ship speed in ice. These however often fail to account for the joint effect of the ice conditions on ship's speed. Moreover, they omit the effect of ice compression. The latter, when combined with the presence of ridges, can significantly limit the capabilities of an ice-strengthened ship, and potentially bring her to a halt, even if the actual ice conditions are within the design range for the given ship.This paper introduces two probabilistic, data-driven models that predict a ship's speed and the situations where a ship is likely to get stuck in ice based on the joint effect of ice features such as the thickness and concentration of level ice, ice ridges, rafted ice, moreover ice compression is considered.To develop the models, two full-scale datasets were utilized. First, the dataset about the performance of a selected ship in ice is acquired from the automatic identification system. Second, the dataset containing numerical description of the ice field is obtained from a numerical ice model HELMI, developed in the Finnish Meteorological Institute.The collected datasets describe a single and unassisted trip of an ice-strengthened bulk carrier between two Finnish ports in the presence of challenging ice conditions, which varied in time and space.The relations between ship performance and the ice conditions were established using Bayesian networks and selected learning algorithms.The obtained results show good prediction power of the models. This means, on average 80{\%} for predicting the ship's speed within specified bins, and above 90{\%} for predicting cases where a ship may get stuck in ice.",
keywords = "Bayesian networks, Machine learning, Ship beset in ice, Ship performance in ice, Bayesian networks, Machine learning, Ship beset in ice, Ship performance in ice, Bayesian networks, Machine learning, Ship beset in ice, Ship performance in ice",
author = "Jakub Montewka and Floris Goerlandt and Pentti Kujala and Mikko Lensu",
note = "VK: T20404",
year = "2015",
doi = "10.1016/j.coldregions.2014.12.009",
language = "English",
volume = "112",
pages = "14--28",
journal = "Cold Regions Science and Technology",
issn = "0165-232X",
publisher = "Elsevier",

}

RIS - Download

TY - JOUR

T1 - Towards probabilistic models for the prediction of a ship performance in dynamic ice

AU - Montewka, Jakub

AU - Goerlandt, Floris

AU - Kujala, Pentti

AU - Lensu, Mikko

N1 - VK: T20404

PY - 2015

Y1 - 2015

N2 - For safe and efficient exploitation of ice-covered waters, knowledge about ship performance in ice is crucial. The literature describes numerical and semi-empirical models that characterize ship speed in ice. These however often fail to account for the joint effect of the ice conditions on ship's speed. Moreover, they omit the effect of ice compression. The latter, when combined with the presence of ridges, can significantly limit the capabilities of an ice-strengthened ship, and potentially bring her to a halt, even if the actual ice conditions are within the design range for the given ship.This paper introduces two probabilistic, data-driven models that predict a ship's speed and the situations where a ship is likely to get stuck in ice based on the joint effect of ice features such as the thickness and concentration of level ice, ice ridges, rafted ice, moreover ice compression is considered.To develop the models, two full-scale datasets were utilized. First, the dataset about the performance of a selected ship in ice is acquired from the automatic identification system. Second, the dataset containing numerical description of the ice field is obtained from a numerical ice model HELMI, developed in the Finnish Meteorological Institute.The collected datasets describe a single and unassisted trip of an ice-strengthened bulk carrier between two Finnish ports in the presence of challenging ice conditions, which varied in time and space.The relations between ship performance and the ice conditions were established using Bayesian networks and selected learning algorithms.The obtained results show good prediction power of the models. This means, on average 80% for predicting the ship's speed within specified bins, and above 90% for predicting cases where a ship may get stuck in ice.

AB - For safe and efficient exploitation of ice-covered waters, knowledge about ship performance in ice is crucial. The literature describes numerical and semi-empirical models that characterize ship speed in ice. These however often fail to account for the joint effect of the ice conditions on ship's speed. Moreover, they omit the effect of ice compression. The latter, when combined with the presence of ridges, can significantly limit the capabilities of an ice-strengthened ship, and potentially bring her to a halt, even if the actual ice conditions are within the design range for the given ship.This paper introduces two probabilistic, data-driven models that predict a ship's speed and the situations where a ship is likely to get stuck in ice based on the joint effect of ice features such as the thickness and concentration of level ice, ice ridges, rafted ice, moreover ice compression is considered.To develop the models, two full-scale datasets were utilized. First, the dataset about the performance of a selected ship in ice is acquired from the automatic identification system. Second, the dataset containing numerical description of the ice field is obtained from a numerical ice model HELMI, developed in the Finnish Meteorological Institute.The collected datasets describe a single and unassisted trip of an ice-strengthened bulk carrier between two Finnish ports in the presence of challenging ice conditions, which varied in time and space.The relations between ship performance and the ice conditions were established using Bayesian networks and selected learning algorithms.The obtained results show good prediction power of the models. This means, on average 80% for predicting the ship's speed within specified bins, and above 90% for predicting cases where a ship may get stuck in ice.

KW - Bayesian networks

KW - Machine learning

KW - Ship beset in ice

KW - Ship performance in ice

KW - Bayesian networks

KW - Machine learning

KW - Ship beset in ice

KW - Ship performance in ice

KW - Bayesian networks

KW - Machine learning

KW - Ship beset in ice

KW - Ship performance in ice

UR - http://www.sciencedirect.com/science/article/pii/S0165232X14002262

U2 - 10.1016/j.coldregions.2014.12.009

DO - 10.1016/j.coldregions.2014.12.009

M3 - Article

VL - 112

SP - 14

EP - 28

JO - Cold Regions Science and Technology

JF - Cold Regions Science and Technology

SN - 0165-232X

ER -

ID: 2018225