MACHINE LEARNING-BASED ANALYSIS OF DRILLING PARAMETERS FOR ROP PREDICTION IN OFFSHORE OPERATIONS: A COMPARATIVE STUDY OF RANDOM FOREST, GRADIENT BOOSTING, AND ADABOOST REGRESSION MODELS

Offshore drilling remains an indispensable pillar of global energy extraction, necessitating intricate evaluations of operational dynamics to enhance both performance and safety.This study delves into the interplay between drilling parameters and the Rate of Penetration (ROP) through sophisticated machine learning methodologies, striving to refine efficiency within offshore drilling activities.Its significance is underscored by an exhaustive assessment of multiple regression frameworks for ROP prediction, yielding crucial insights to aid operational decisionmaking in intricate drilling landscapes.The methodological framework incorporates three distinct machine learning paradigms-Random Forest, Gradient Boosting, and AdaBoost Regression-applied to the Drilling Log Dataset from well number 15/9-F-15, situated within the Volve Oil Field of the North Sea.The dataset fuses real-time Divya Soundarapandian https://iaeme.com/Home/journal/IJARET