The main objective of this work is to use machine-learning (ML) algorithms to develop a powerful model to predict well-integrity (WI) risk categories of gas-lifted wells. The model described in the complete paper can predict well-risk level and provide a unique method to convert associated failure risk of each element in the well envelope into tangible values.

The model can be subdivided into four submodels as follows:

This classification was performed using ML algorithms such as logistic regression, decision trees, random forest, Gaussian, K-nearest neighbor, and support-vector machine classifiers.

Steps of Building the Model.

The Journal of Petroleum Technology, the Society of Petroleum Engineers’ flagship magazine, presents authoritative briefs and features on technology advancements in exploration and production, oil and gas industry issues, and news about SPE and its members.

ISSN: 1944-978X (Online) ISSN: 0149-2136 (Print)