Open AccessSelection of optimal forecasting models in the navigation safety decision support system
Author(s) -
V V Astrein,
S I Kondratyev,
A L Boran-Keshishyan
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/2061/1/012104
Subject(s) - reliability (semiconductor) , relation (database) , computer science , block (permutation group theory) , artificial neural network , decision support system , relevance (law) , selection (genetic algorithm) , process (computing) , operations research , control (management) , probabilistic forecasting , artificial intelligence , data mining , engineering , mathematics , power (physics) , physics , geometry , quantum mechanics , probabilistic logic , political science , law , operating system
One of the most important components of the navigation safety decision support system (NS DSS) is the forecasting block. The results of his work have an impact on the type of control actions formed by the system. The effectiveness of this block depends on the forecasting methods used. The accuracy of the result of the forecasting block depends largely on the choice of the forecasting method.To develop reliable forecasts, it is necessary to determine forecasting methods, in relation to the specifics of the functioning of the NS DSS. Several ways of solving the problem of choosing the appropriate forecasting method in relation to the NS DSS are proposed. Two methods are demonstrated: Artificial neural networks (ANN) using precedents and Intelligent process analysis (IPA). Solving the problem of choosing the optimal method will guarantee obtaining a forecast with a certain level of accuracy, which will significantly increase the reliability of the forecast and, as a result, the effectiveness of the NS DSS. Solving the problem of choosing the optimal method will guarantee obtaining a forecast with a certain level of accuracy, which will significantly increase the reliability of the forecast and, as a result, the effectiveness of the NS DSS, which allows us to assert the relevance of this problem and further research.