Publications

New Approach for Optimizing Vertical Well Placement through Reservoir Opportunity Index and Naïve Bayes Classifier Methods at Multi-Tank Reservoir in Lembak Field South Sumatera

Proceedings Title : Proc. Indon. Petrol. Assoc., 48th Ann. Conv., 2024

The multi-tank reservoir is one of the most challenging reservoir types for placing infill and step-out wells in the development phase either for green or brownfields. In this type of reservoir, we also deal with facies heterogeneity which can cause a dry hole or shale out reservoir when the target sand is drilled. Based on the experience we have had, we propose a method to overcome this problem with ROI (Reservoir Opportunity Index) and Naïve Bayes Classifier.

The Reservoir Opportunity Index (ROI) constitutes a sophisticated metric within the realm of petroleum geoscience and reservoir engineering, designed to holistically evaluate the potential for reservoir development and exploitation within a specific geological location. This metric incorporates a nuanced analysis of diverse factors, encompassing geological attributes, hydrocarbon prospects, and economic feasibility. Through a systematic integration of these variables, the ROI serves as an invaluable analytical tool for stakeholders such as energy corporations and investors, facilitating a comprehensive assessment of the attractiveness of a reservoir for exploration and production endeavors.

In reservoir engineering, Naive Bayes Classification serves as a powerful tool for optimizing well placement strategies. Leveraging probabilistic models, Naive Bayes analyzes historical data on reservoir properties, such as porosity, permeability, and fluid saturation, to classify potential well locations based on their likelihood of success. This approach assumes independence among the input features, simplifying the complexity of the reservoir system. By integrating data from past drilling experiences and subsurface geology, Naive Bayes helps identify patterns and relationships that can guide engineers in making informed decisions regarding optimal well placement. The algorithm's ability to swiftly process large datasets and provide real-time predictions enables reservoir engineers to enhance their decision-making process, ultimately leading to more efficient and cost-effective well placement strategies that maximize hydrocarbon recovery from the subsurface reservoirs.

Log In as an IPA Member to Download Publication for Free.