Pengembangan Simulasi Penghindaran Dini Tabrakan Kapal dengan Metode Fuzzy Inference Berbasis Collision Risk Index

Pratama, Fernanda Wahyu (2024) Pengembangan Simulasi Penghindaran Dini Tabrakan Kapal dengan Metode Fuzzy Inference Berbasis Collision Risk Index. Masters thesis, Institut Teknologi Sepuluh Nopember.

Text 6018221001-Master_Thesis.pdf - Accepted Version Download (19MB)

Abstract

Tabrakan kapal dapat terjadi karena perilaku pengoperasian yang berbahaya (human errors). Saat kapal berpapasan dengan kapal lain, nakhoda akan melihat kondisi sekitar kapal dengan radar, Automatic Identification System (AIS), atau juga dengan mata nakhoda sendiri untuk mendapatkan informasi posisi dan informasi perjalanan kapal lainnya. Dengan mengandalkan peralatan navigasi ataupun pengalaman dari nakhoda, terdapat beberapa kekurangan yang perlu diperhatikan. Oleh karena itu, simulasi penghindaran dini tabrakan kapal sangat diperlukan, untuk mengambarkan perilaku kapal sesuai dengan perintah nakhoda berdasarakan hasil nilai collision risk (CR). Tesis ini berfokus pada simulasi penghindaran tabrakan kapal untuk meningkatkan safety awarness awak kru kapal dalam kondisi perairan terbatas (restricted waterway) seperti cannal, jalur masuk pelabuhan, dll. Simulasi dikembangkan dengan menggabungkan evaluasi risiko tabrakan berbasis metode numerik, menggunakan parameter seperti Distance of Closest Point of Approach (DCPA), Time to Closest Point of Approach (TCPA), dan domain kapal untuk memberikan penilaian yang lebih akurat dan efektif dalam menghindari tabrakan. Dalam simulasi ini, kapal dipertimbangkan mempertahankan kecepatannya karena masih dalam zona aman dalam melakukan penghindaran tabrakan. Penelitian ini mengembangkan model algoritma evaluasi risiko tabrakan kapal berbasis fuzzy inference yang menggabungkan indeks risiko tabrakan, warning risk by ship domains, dan closest the point of approach dengan prediksi gerakan kapal berbasis model matematika manuver kapal. Data radar dan Automatic Identification System (AIS) kapal digunakan untuk memprediksi perubahan posisi kapal, yang kemudian diproses melalui model matematika untuk menghasilkan penilaian risiko tabrakan. Tiga variasi bentuk domaian digunakan dalam penelitian ini. Hasil simulasi menunjukkan bahwa kapal dapat menghindari tabrakan dengan baik, dengan domain elips dan telur memberikan penilaian CR yang lebih baik dibandingkan domain lingkaran yang kurang variatif dalam ukuran.
========================================================================================================================
When ships encounter each other, the captain uses radar, the Automatic Identification System (AIS), or visual observation to gather positional and navigational information of other vessels. Relying solely on navigational equipment or the captain's experience has its limitations. Therefore, early collision avoidance simulations are crucial to depict the ship's behavior according to the captain's commands based on collision risk (CR) values. This thesis focuses on collision avoidance simulations to enhance crew safety awareness in restricted waterways such as canals and port entry routes. The simulation is developed by integrating numerical method-based collision risk evaluation, using parameters like Distance of Closest Point of Approach (DCPA), Time to Closest Point of Approach (TCPA), and ship domains to provide a more accurate and effective collision avoidance assessment. In this simulation, the ship is considered to maintain its speed as it is still within the safe zone for collision avoidance. This research develops a ship collision risk evaluation algorithm based on fuzzy inference, combining collision risk indices, warning risk by ship domains, and the closest point of approach with ship movement prediction based on a mathematical maneuvering model. Radar and Automatic Identification System (AIS) data are used to predict changes in ship positions, which are then processed through the mathematical model to produce collision risk assessments. Three variations of domain shapes are used in this study. The simulation results show that ships can effectively avoid collisions, with elliptical and egg-shaped domains providing better CR assessments compared to circular domains, which are less variable in size.

Item Type:	Thesis (Masters)
Uncontrolled Keywords:	Metode fuzzy logic, Risiko tabrakan, Domain kapal, Metode closest point of approach, Prediksi gerak kapal, Penghindaran tabrakan, Fuzzy logic method, Collision risk, Ship domian, Closest point of approach method, Ship movement prediction, Collision avoidance.
Subjects:	T Technology > T Technology (General) > T174.5 Technology--Risk assessment. T Technology > T Technology (General) > T57.62 Simulation V Naval Science > VM Naval architecture. Shipbuilding. Marine engineering > VM365 Remote submersibles. Autonomous vehicles.
Divisions:	Faculty of Marine Technology (MARTECH) > Naval Architecture and Shipbuilding Engineering > 36101-(S2) Master Thesis
Depositing User:	Fernanda Wahyu Pratama
Date Deposited:	10 Aug 2024 15:30
Last Modified:	10 Aug 2024 15:30
URI:	http://repository.its.ac.id/id/eprint/114235

Actions (login required)

View Item