ASSESSING POTENTIAL COLLISION RISK OF MARITIME AUTONOMOUS SURFACE SHIPS IN INDONESIA USING BAYESIAN NETWORKS

Prakoso, Epifanius Kristio (2025) ASSESSING POTENTIAL COLLISION RISK OF MARITIME AUTONOMOUS SURFACE SHIPS IN INDONESIA USING BAYESIAN NETWORKS. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5019211033-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 5 August 2025. Download (3MB) | Request a copy

Abstract

Perkembangan pesat Maritime Autonomous Surface Ships (MASS) menghadirkan tantangan baru, terutama dalam menjamin keselamatan navigasi dan mencegah tabrakan. Penelitian ini bertujuan untuk menilai potensi risiko tabrakan dari operasi MASS di perairan Indonesia menggunakan pendekatan Bayesian Network (BN). Model dibangun berdasarkan studi literatur, penilaian ahli, serta dukungan kuesioner yang ditujukan kepada pelaut profesional, dan mencakup variabel dari aspek teknis, lingkungan, dan manajerial. Faktor risiko utama seperti navigasi lingkungan, komunikasi kapal-darat, kegagalan peralatan, dan kinerja manajemen dimodelkan dan dianalisis. Hasil menunjukkan probabilitas tabrakan yang signifikan sebesar 30%, yang menandakan tingkat risiko menengah. Namun, penelitian ini menghadapi keterbatasan seperti tidak tersedianya data historis mengenai operasi MASS di Indonesia dan kesulitan mengumpulkan respons kuesioner dari pelaut karena pekerjaan mereka di laut lepas dan keterbatasan koneksi. Keterbatasan ini diatasi melalui masukan ahli dan perbandingan dengan literatur internasional. Meskipun menghadapi tantangan, model Bayesian ini memberikan kerangka kerja awal yang berguna untuk memahami risiko tabrakan dan menjadi acuan pengembangan serta regulasi MASS di Indonesia. Penelitian lanjutan disarankan untuk melibatkan lebih banyak ahli dan pengujian empiris seiring berkembangnya teknologi MASS di masa mendatang.==============================================================================================================The rapid development of Maritime Autonomous Surface Ships (MASS) brings new challenges, particularly in ensuring safe navigation and preventing collisions. This research aims to assess the potential collision risks of MASS operations in Indonesian waters using a Bayesian Network (BN) approach. The model was built based on literature studies, expert judgment, and a supporting questionnaire targeting professional seafarers, and includes variables from technical, environmental, and management domains. Key risk factors such as environmental navigation, ship-shore communication, equipment failure, and management performance were modeled and analyzed. Results show a significant collision probability of 30%, indicating moderate level of risk under certain conditions in several individual nodes. However, the research also faced limitations, such as the lack of historical data on MASS operations in Indonesia and difficulty in collecting questionnaire responses from seafarers due to their offshore duties and limited connectivity. These constraints were mitigated through expert input and cross-referencing with international literature. Despite these challenges, the Bayesian model provides a useful early-stage framework for understanding collision risk and guiding future development and regulation of MASS in Indonesia. Further research is recommended to incorporate a broader expert base and empirical testing as MASS technology becomes more widely implemented.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Kapal otonom, Bayesian Network, risiko tabrakan===============================================================================================================Autonomous ships, Bayesian Network, collision risk
Subjects:	T Technology > T Technology (General) > T174.5 Technology--Risk assessment.
Divisions:	Faculty of Marine Technology (MARTECH) > Marine Engineering > 36202-(S1) Undergraduate Thesis
Depositing User:	Epifanius Kristio Prakoso Putra
Date Deposited:	04 Aug 2025 08:27
Last Modified:	04 Aug 2025 08:27
URI:	http://repository.its.ac.id/id/eprint/126921

Actions (login required)

View Item