Artificial Neural Network Model for Risk-Based Inspection Screening Assessment of Oil and Gas Production System

摘要

Screening assessment is a part of risk-based inspection (RBI)methodology, which enables the filtering of equipment that makes asignificant contribution to the overall risk of the system. The screeningassessment directs detailed assessment to focus on higher-risk items;thus, resources (e.g., time and labor) can be allocated more effectivelyand efficiently. However, performing RBI screening assessmentrequires considerable time and resources, and its results are prone tohigh variability, due to its qualitative nature. To mitigate theaforementioned challenges and to make the knowledge work involvedlean (i.e. to minimize waste), an artificial neural network (ANN) modelis suggested for performing the RBI screening assessment. Thedevelopment of the ANN model is demonstrated by using a datasetcontaining the data and information from an RBI assessment conductedfor onshore and offshore hydrocarbon production and process systems.It is revealed that the suggested model is capable of achievingrespectable performance with 90.65% accuracy, 82.76% precision, and76.59% recall.