Application of an Advanced Data Analytics Methodology to Predict Hydrocarbon Recovery Factor Variance Between Early Phases of Appraisal and Post-Sanction in Gulf of Mexico Deep Offshore Assets

Abstract

Abstract Estimated Ultimate Recovery (EUR) is the expected amount of hydrocarbons that can ultimately be produced from a reservoir or group of reservoirs (oilfield) given certain subsurface characteristics and technologies. Usually, the EUR is estimated at different phases during the lifecycle of an E&P asset for a continuous economic assessment. Reduction in uncertainity in EUR estimation due to acquisition and analysis of data during early phases of exploration and appraisal would enable effective decision making process during later phases of development and production. Being able to predict EUR effectively early on in the field lifecycle would improve project viability and enable higher quality investment decisions during latter phases. In this study, EUR variance (?EUR) is computed between the EUR at pre-sanction and the EUR at post-sanction after first oil production. EUR variance is frequently overestimated as several reservoir parameters are unknown or less certain during appraisal phase. This paper presents a novel data-driven methodology to accurately determine ?EUR using known EUR at appraisal phase and data associated to oilfield characteristics, complexity and definition level. The methodology employs multi-variate statistical techniques and optimization procedure to construct a robust EUR estimation model and to identify key drivers that impact ?EUR. The analytical methodology was successfully validated and tested using real data from over 50 large operating assets of the Deepwater Gulf of Mexico (GOM) oilfields. Numerical case studies show the methodology was able to satisfactorily predict ?EUR and potential project train-wrecks with 73 and 85 percent of accuracy, respectively, as well as to identify, rank, and classify those key parameters responsible for both under- and over-estimating the EUR. The proposed model creates opportunities to reduce "appraisal gaps" between pre-sanction and post-sanction EUR and improve the field development planning process in a cost-effective manner. This data-driven methodology has immense potential in deepwater oilfields and other hydrocarbon plays to improve field appraisal and development strategy and also assess other risks in hydrocarbon management.