首页|Machine Learning With Multi-Source Data to Predict and Explain Marine Pilot Occupational Accidents

Machine Learning With Multi-Source Data to Predict and Explain Marine Pilot Occupational Accidents

扫码查看
原文链接
  • NETL
  • NSTL
  • Natl Taiwan Ocean Univ
Marine pilot occupational accidents during transfer to/from ships are the primary concern of the International MarinePilots’ Association (IMPA) and industry professionals. There are multiple transfer methods for marine pilots, with themost common being the pilot boat. To reach the mother ship bridge, the following stages must be safely completed: cartransfer, walking on the pier, pier to pilot boat, pilot transfer by boat, cutter to pilot ladder, mother ship freeboardclimbing, and ship deck to the bridge. Each stage has its own risk. Previous accident records and expert opinions arecommonly used to conduct a risk analysis and take preventive actions. However, the reports vary in scope and are oftencomplex, making qualitative analysis a timeintensive task. To overcome this challenge, this study aggregates 500 reportsto create a multi-source dataset describing instances of undesired events. A ML (machine learning) approach is used topredict and explain marine pilot occupational accidents. Analyzing the importance of factors distinguishing betweenaccidents, incidents, and non-compliance, we conclude that workplace factors are more dangerous than environmentalfactors. The findings of this study provide a foundation for developing a unified accident reporting system for predictingaccidents on a wider scale.

Marine pilotPilot ladderOccupational accidentRF (random forest)Explainable ML

Gokhan Camliyurt、Youngsoo Park、Daewon Kim、Won-Sik Kang、Sangwon Park

展开 >

Department of Navigation, Graduate School, Korea Maritime, and Ocean University, South Korea

Division of Navigation Convergence Studies, Korea Maritime, and Ocean University, South Korea

College of Ocean Sciences, Jeju National University, South Korea

Logistics and Maritime Industry Research Department, Korea Maritime Institute, South Korea

展开 >

2023

10.51400/2709-6998.2709

Journal of Marine Science and Technology

Journal of Marine Science and Technology

ISSN：1023-2796
年,卷(期)：2023.31(4)
  • 1
  • 56