LITERATURE REVIEW

Sand production remains one of the most persistent challenges in oil and gas operations, particularly in unconsolidated sandstone reservoirs where weak formations are prone to failure under changing pressure and stress conditions. This study explores the advancements in sand prediction and management techniques, focusing on the integration of artificial intelligence tools such as Artificial Neural Networks (ANN) and Support Vector Machines (SVM) for enhanced predictive accuracy. By analyzing key reservoir parameters, machine learning models were developed to classify wells based on their sand production potential. The research compares the predictive performance of ANN and SVM algorithms, identifying the most reliable and adaptable approach for field applications, especially in datascarce environments. Findings from this study contribute to improved decisionmaking in sand control strategy selection, reduced equipment damage, minimized production losses, and more sustainable well management practices in unconsolidated reservoirs.