Revolutionizing Resource-to-Reserve with COREAI: Unlocking New Opportunities and Accelerating Success

Abstract This paper presents the development and application of COREAI, a machine-learning-based platform designed to manage and analyze Malaysia's extensive legacy petroleum engineering core data. The platform transforms data from 23,000 core plugs across 608 wells into predictive models that estimate rock and fluid properties, validate subsurface analyses, and reduce uncertainty. COREAI has been deployed in 25 fields, achieving significant cost savings, enhancing reservoir modeling, and optimizing field development, particularly for marginal fields. This study demonstrates how digitization and advanced analytics can unlock the value of legacy data, providing accurate and timely insights for subsurface evaluations. The objective of this project is to develop an intelligent platform using Azure web applications to manage and analyze Malaysia's extensive legacy petroleum engineering core data. Figure 1 shows the regional availability of core analysis data in Malaysia. By integrating machine learning, the platform predicts rock and fluid properties, validates subsurface analyses, and reduces uncertainty. These 23,000 core plugs and associated analyses is transformed into a robust reservoir properties knowledge model, enhancing reserve estimation accuracy, optimizing field development, and significantly reducing costs and data-related inefficiencies.