作者: Pentti Kujala , Hilary Sinclair , Jakub Montewka , Jari Haapala , Mikko Lensu
DOI:
关键词: Probabilistic logic 、 Ice navigation 、 Ice thickness 、 Geography 、 Marine engineering 、 Joint (geology) 、 Bayesian network
摘要: Navigation in ice has received substantial attention over recent decades. This increased led to the development of numerical and semi-empirical models that characterize ship performance ice. These consider numerous parameters such as; level thickness, ridged thickness concentration additive fashion. However, they fail account for joint effect above features on speed. Moreover, compression is usually omitted. paper introduces probabilistic models, based field observations, predict a ship’s speed situations where probable get stuck selected features, as ice, ridges, rafted compression. To develop Bayesian Belief Network used. The case study presented this considers single unassisted trip an ice-strengthened bulk carrier between two Finnish ports presence challenging conditions obtained results show very good prediction power models.