Prediksi Trajectory Kapal dan Critical Collision Zone Di Selat Sunda Berbasis Data Automatic Identification System dan Cuaca

Arum, Kanugrahing Christy Sekar (2024) Prediksi Trajectory Kapal dan Critical Collision Zone Di Selat Sunda Berbasis Data Automatic Identification System dan Cuaca. Diploma thesis, Institut Teknologi Sepuluh Nopember.

Text 2043211038-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 April 2027. Download (7MB) | Request a copy

Abstract

Indonesia sebagai negara maritim terbesar di dunia memiliki aktivitas maritim yang padat dimana dapat menyebabkan peningkatan risiko kecelakaan kapal. Kondisi cuaca ekstrem juga berpengaruh besar terhadap keselamatan pelayaran. Adanya cuaca ekstrem di laut seperti badai dan angin kencang dapat meningkatkan risiko kecelakaan kapal. Salah satu wilayah yang paling berisiko adalah Selat Sunda sebagai salah satu jalur pelayaran strategis yang sering dilalui kapal lokal dan internasional. Penelitian ini bertujuan untuk mengidentifikasi titik-titik yang berisiko tinggi terjadinya tubrukan kapal di Selat Sunda atau yang dapat disebut Critical Collision Zone (CCZ). CCZ ditentukan berdasarkan hasil analisis prediksi trajectory kapal menggunakan Bidirectional Gated Recurrent Unit (Bi-GRU) dan clustering menggunakan Density-Based Spatial Clustering of Applications with Noise (DBSCAN). Prediksi trajectory kapal diperoleh dari data Automatic Identification System (AIS) dan cuaca. Data AIS pada setiap kapal memberikan informasi penting mengenai posisi, kecepatan, dan arah kapal secara real-time, yang jika diintegrasikan dengan data cuaca dapat membentuk trajectory kapal yang lebih akurat dan membantu dalam meningkatkan keamanan pelayaran. Setelah CCZ teridentifikasi, peluang terjadinya tubrukan kapal di dalam CCZ dihitung menggunakan Monte Carlo Simulation (MCS). Hasil penelitian menunjukkan bahwa mayoritas kapal yang melintas adalah Bulk Carrier dengan negara asal keberangkatan terbanyak dari Indonesia dan Panama. Model prediksi dengan data cuaca memberikan performa lebih baik dalam identifikasi CCZ dan estimasi peluang tubrukan, dengan nilai MAE dan MSE yang lebih rendah serta silhouette coefficient yang lebih tinggi, menunjukkan peningkatan akurasi dalam mengidentifikasi wilayah berisiko dan estimasi peluang tubrukan. ==================================================================================================================================
Indonesia as the largest maritime country in the world has dense maritime activities which can lead to an increased risk of ship accidents. Extreme weather conditions also have a major effect on shipping safety. The presence of extreme weather at sea such as storms and strong winds can increase the risk of ship accidents. One of the most risky areas is the Sunda Strait as one of the strategic shipping lanes that is often traveled by local and international ships. This study aims to identify points that are at high risk of ship collisions in the Sunda Strait or what can be called the Critical Collision Zone (CCZ). CCZ is determined based on the analysis of ship trajectory prediction using Bidirectional Gated Recurrent Unit (Bi-GRU) and clustering using Density-Based Spatial Clustering of Applications with Noise (DBSCAN). Ship trajectory prediction is obtained from Automatic Identification System (AIS) and weather data. The AIS data on each ship provides important information about the ship's position, speed, and course in real-time, which when integrated with weather data can form a more accurate ship trajectory and help in improving shipping safety. Once the CCZ is identified, the probability of a ship collision within the CCZ is calculated using Monte Carlo Simulation (MCS). The results indicate that the majority of passing vessels are Bulk Carriers, with the most common departure countries being Indonesia and Panama. The predictive model using weather data showed better performance in identifying CCZs and estimating collision probability, with lower Mean Absolute Error (MAE) and Mean Squared Error (MSE) values as well as higher silhouette coefficient, indicating an improvement in the accuracy of identifying high-risk areas and estimating collision probabilities.

Item Type:	Thesis (Diploma)
Uncontrolled Keywords:	Automatic Identification System (AIS), Critical Collision Zone (CCZ), Cuaca, Trajectory Kapal, Tubrukan Kapal, Ship Collision, Ship Trajectory, Weather
Subjects:	H Social Sciences > HA Statistics > HA31.7 Estimation H Social Sciences > HE Transportation and Communications > HE5614.3.N5 Traffic accidents Q Science > QA Mathematics > QA278.55 Cluster analysis Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science)
Divisions:	Faculty of Vocational > 49501-Business Statistics
Depositing User:	Kanugrahing Christy Sekar Arum
Date Deposited:	31 Dec 2024 07:44
Last Modified:	31 Dec 2024 07:44
URI:	http://repository.its.ac.id/id/eprint/116084

Actions (login required)

View Item