Data science application in structural integrity analysis of fixed offshore jacket platform

Abstract

Accuracy in analysing the integrity of a structure is critical for determining the structure's fitness for service and reliability status. Today, a variety of techniques and approaches are applied, including the use of data science applications. Data science is a synthesis of computer science, mathematics, and statistics. Meanwhile, the integrity of a structure is susceptible to a mix of statistical and technical design uncertainties that may remain flexible as long as the structure is capable of successfully managing the encountered load. Numerous applications are used in the oil and gas sector to estimate the probability of failure (POF), but they all have a particular restriction. Integral interference equations based on load versus strength are reliable for determining the POF of fixed offshore structures. This study is a quantitative risk assessment, emphasising the Python application, an improved and reliable method for calculating the POF value. A representative sample of the monopod offshore structure was chosen and subjected to global non-linear analysis in this study. The most reliable form of distribution was predetermined, and the algorithm created using Python was used to apply and compute the suitable integral equation depending on the load and strength conditions. The Python method's result demonstrated a high degree of confidence in calculating the new POF in intact condition from a design perspective, inspection interval, and risk to consider.