МЕТОД АВТОМАТИЗОВАНОЇ КЛАСИФІКАЦІЇ НАДЗВИЧАЙНИХ СИТУАЦІЙ ІЗ СУДНОМ В АКВАТОРІЇ МОРСЬКОГО ПОРТУ

Abstract

The aim of the study is to develop a method that allows to improve the accuracy of classification of emergencies with a ship in the seaport area. To achieve the research objective, an improved method for automated classification of ship emergencies in the seaport area under conditions of deterministic uncertainty is proposed. The method consists of methods for formalising and processing knowledge on determining the classes of ship emergencies. The method of knowledge formalisation is based on the formation of a set of factors influencing the classification of ship emergencies in the seaport water area based on fuzzy binary relations of non-strict preference. The method of knowledge processing is based on the formation of productive rules for classifying emergencies with a ship in the seaport area according to the predicted or current values of the desired factors in a fuzzy formulation. The determination of the membership functions of several fuzzy variables to linguistic variables is based on the processing of expert data represented by a matrix of binary relations of the values of the membership function of the elements of the domain of determining linguistic variables. A comparative assessment of the classes of recognisable emergencies with a vessel in the seaport water area involves solving a multi-criteria optimisation problem using the hierarchy analysis method. As a mathematical model for determining the classes of emergency situations with a vessel in the seaport waters, the article substantiates a logical-linguistic productive hierarchical model. The process of determining the classes of emergency is described with the help of an algebraic model, which is closest to the linguistic description. In case of inexpediency of synthesis of products, it is proposed to use the method of fuzzy identification to reduce the number of product rules. The proposed method allows to improve the accuracy of classification of emergencies with a ship in the seaport area under conditions of deterministic uncertainty. Keywords: classification, maritime port, hierarchy analysis, fuzzy identification, logicallinguistic production hierarchical model, vessel, class of emergency situations, fuzzy variable, linguistic variable