Maritime Weather Prediction Using Fuzzy Logic in Java Sea for Shipping Feasibility

Abstract

Fuzzy logic has mostly been used in different kind of field either in taking decision or control system and forecasting. Fuzzy logic is one that can represent a complex situation into a simple form in a language that is easily caught by humans. Similarly, fuzzy logic can represent the nature of weather to be easily understood by ordinary people as the result of weather forecasting based on meteorological data. In this study, some of the statistical methods were used for weather forecasting, i.e. Auto Regressive (AR), Integrated Moving Average (ARIMA) and some improvement of shortcomings of these traditional methods. The numbers of historical data needed in weather forecasting are not small to avoid low prediction accuracy results in a complex condition. This paper describes the strategy in maritime weather forecasting by means of fuzzy logic. Case studies were conducted in the Java Sea to determine the shipping feasibility in cruise lines Surabaya- Banjarmasin as two major ports that connect Java and Kalimantan (Borneo) island. The results show that the level of prediction accuracy is 86.64% of the Java Sea and prediction accuracy decrease for the next 24 hours. Predictors can demonstrate the ability to detect anomalous high sea waves in the waters of the Java Sea at three locations between Surabaya waters and Banjarmasin waters. This prediction results are proven to be able to determine the shipping feasibility.