A Big Data Analytics Method for the Evaluation of Ship - Ship Collision Risk reflecting Hydrometeorological Conditions

• DOI: 10.1016/j.ress.2021.107674
• Publication Date: 2021-04-13T20:16:35Z
• Authors: Mingyang Zhang; Jakub Montewka; Teemu Manderbacka; Pentti Kujala; Spyros Hirdaris
• Keywords: Automatic Identification System; Identification; Hydrometeorology

This paper presents a big data analytics method for the evaluation of ship-ship collision risk in real operational conditions. The approach makes use from Automatic Identification System (AIS) and nowcast corresponding to time-dependent traffic situations hydro-meteorological conditions respectively. An Avoidance Behavior-based Collision Detection Model (ABCD-M) is introduced identify potential scenarios Risk Indices (CRIs) are quantified when evasive actions taken each detected scenario various voyages. applied on Ro-Pax ships operating over 13 months ice-free period Gulf Finland. Results indicate that estimates may be extremely diverse among voyages, 97.5% triggered only at 45% or more its maximum value. overall CRI given area tends lower adverse It therefore concluded proposed assist with (1) identification critical voyages not currently accounted by existing accident databases, (2) definition commonly agreed criteria set off alarms, (3) estimation profile life cycle fleet operations.