Abstract
In the Oil and Gas (O&G) industry, maximizing production from existing wells is crucial to staying competitive. Identifying and accessing Behind Casing Opportunities (BCO) for Production Enhancement and Idle Well Reactivation (PE-IWR) initiative, provides a cost-effective alternative than drilling new wells to increase the recovery and extend the life of a well. However, accurately predicting BCO in mature offshore fields, particularly in Field S located in offshore East Malaysia, remains a formidable challenge due to complex geological formations, remote locations, and aging infrastructure which have resulted in ambiguous logging results and high operational costs that complicate data collection and reduce the reliability of conventional logging methods.
The purpose of this paper is to discuss an innovative approach to predict BCO by leveraging Artificial Intelligence and Machine Learning (AI-ML) as a pre-screening tool for identifying PE-IWR candidates, showcasing a comprehensive analysis and method to validate the accuracy of the result predictions using Field S as case study. The BCO identification solution has expedited the discovery of multiple promising candidates, from previously untapped areas, resulting in a swift realization of its benefits and has made the decision-making process much easier. One case that stands out is the discovery of promising BCO in Well-01 of Field S, which indicated significant potential oil gain. A collaborative effort between the engineers and Subject Matter Experts (SMEs) was initiated, with the assistance of the solution, to quickly capitalize on this opportunity that involved comprehensive subsurface technical assessment, including detail analysis of well logs, investigation of well intervention and integrity history, evaluation of reservoir performance, and review of well production behavior.
The successful implementation in Field S, marked another favorable outcome that demonstrates the effectiveness and practicality of the AI-ML solution in enhancing BCO identification to optimize production performance and increase profitability. These implementations not only empower engineers to uncover hidden reservoir potentials but also foster a proactive approach to well and reservoir management and surveillance through real-time data driven approach.
