Bayesian Updating for Probability of Failure of Jacket Platforms in Malaysia

Abstract- The Jacket platform codes such as API LRFD and ISO
19902 are based on probabilistic design of component and joint
reliability. They consider overall structural integrity,
redundancy and multiple failure paths only indirectly by using
structural integrity assessment methods. In this paper,
probability of failure is determined as per design requirement of
100 year extreme conditions using Monte Carlo simulation. To
get information on maximum strength, maximum wave height
was increased till the reserve strength ratio reached 1, using
SACS pushover analysis. Stokes’s 5th order theory and
Morrison Equation were used for finding the environmental
loads. Regression analysis was used for the load model using
surface fit tool of Matlab. The wave which gave an RSR value of
1 is considered as the maximum wave, the jacket can withstand
with the available resistance of material. This theory has already
been applied on land based structures, such as proof loading used
against existing structures to gauge the strength of structure.
This new maximum wave height has been used to find the
updated failure probability and compare it with failure
probability of design wave. The study covers one platform, and
recommendation is made whether the platform is suitable for the
extension of life or not. This study can further lead to updating
based on new information on material resistance of platforms.
Keywords: Jacket platform; Probability of failure; Environmental
loading; reserve strength ratio; Bayesian updating