FORECASTING THE NUMBER OF SEARCH AND RESCUE OPERATIONS FOR SHIP ACCIDENTS IN INDONESIA USING FOURIER SERIES ANALYSIS (FSA)

Abstrak

As an archipelago country, Indonesia is a national and international route. This position makes high ship mobility which also increases the risk of ship accidents. To address this issue, based on these conditions, a prediction is required to forecast ship accidents in Indonesia for the upcoming period using an effective method. Through data forecasting, we can map the readiness of Basarnas resources in conducting search and rescue operations for ship accidents. Forecasting data for search and rescue operations in ship accidents is important because it can predict the quantity of needed search and rescue operations. These can be effective measures to reduce casualties in accidents of this type. This research uses the Fourier Series Analysis (FSA) method, which doesn’t require parametric assumption. Additionally, the FSA method can be used for data with unknown patterns. The data used is divided into training data and testing data. The training data used in this research is the number of search and rescue operations from January 2021 to December 2022, while the testing data is from January 2023 to December 2023. The analysis results of this study indicate that forecasting using the FSA method has a MAPE of 25.758%, which falls into the category of reasonable forecasting accuracy and with an optimal and a GCV of 166.586. The results of future predictions are in the form of a mathematical model that can be used by entering the time variable that you want to predict. The anticipated benefits of this research are to contribute to Basarnas’s planning and execution of search and rescue operations for shipwrecks, enrich academic literature on forecasting methodologies, and enhance public awareness of search and rescue operations in Indonesia