Drilling stuck pipe classification and mitigation in the Gulf of Suez oil fields using artificial intelligence

Abstract

Abstract Developing a reliable classification model for drilling pipe stuck is crucial for decision-makers in the petroleum drilling rig. Artificial intelligence (AI) includes several machine learning (ML) algorithms that are used for efficient predictive analytics, optimization, and decision making. Therefore, a comparison analysis for ML models is required to guide practitioners for the appropriate predictive model. Twelve ML techniques are used for drilling pipe stuck such as artificial neural networks, logistic regression, and ensemble methods such as scalable boosting trees and random forest. The drilling cases of the Gulf of Suez wells are collected as an actual dataset for analyzing the ML performance. The key contribution of the study is to automate pipe stuck classification using ML algorithms and mitigate the pipe stuck cases using the genetic algorithm optimization. Out of 12 AI techniques, the results presented that the most reliable algorithm was extremely randomized trees (extra trees) with 100% classification accuracy based on testing dataset. Moreover, this research presents a public open dataset for the drilled wells at the Gulf of Suez to be used for the future experiments, algorithms’ validation, and analysis.