Akademik Veri Yönetim Sistemi
Naval Engineers Journal, sa.133, ss.139-156, 2021 (SCI-Expanded)
Abstract
e systems on the ship have various uids in terms of their
physical and chemical properties. e transfer and storage
of these uids are carried out with the help of pipeline
systems. Although maintenance is carried out periodically
a er the rst installation, it is impossible to see what is
happening inside the pipe systems. For this reason, the
risk levels of these systems should be analyzed so as not to
be faced with hidden hazardous situations that may arise
suddenly. Although the design of the ship system is within
the framework of the rules and regulations, practical
knowledge is required in the operational process. ese
experienced data can be easily obtained from the personnel
who worked on the ships, which means that they can directly
reach the source.
In this study, fuzzy logic is used in the data collection
process, and the Bayesian network structure is organized
to analyze the hazard probability of pipeline failures.
Vibration, maintenance breakdowns, and improper
handling of pipelines are the leading causes of pipeline
failure. e importance of these is obtained from the
result values given by the modeled Bayesian Network.
is research includes the root causes that trigger pipeline
failures, and these reasons have been assessed in terms of
their level of impact.