Eradicating Biases and Establishing Consistency in Geological Chance of Success

Abstract

Geoscientists are bound to have a degree of bias based on their own knowledge, experience, perception, adversity to risk, education, or pre-conceived beliefs. Such subjectivity may lead to a prejudice in making a decision unless this is properly recognized and corrected. As such, this may result in a distorted view of the likelihood on a decision to ‘drill-or-drop,’ if these pre-drill probability predictions are not rationalized. It is therefore extremely important to improve probability assessments by undertaking different approaches, such as setting up detailed and consistent protocols, company-wide standardization or by applying specific elicitation methods. A statistical analysis was undertaken using pre- and post-drill Geological Chance of Success (gCOS) and P-mean volume of the prospects that were drilled vis-a-vis prospects yet to be drilled. The reason for this is to identify the range of the pessimistic and/or optimistic evaluations by the risk reviewers. The purpose then becomes to derive a more stringent and authentic method by which such high deviations in risk estimations, and consistency with the methodology for prospective resource estimations, could be minimized with any potential biases removed. A historical database from the company's assets, spanning over a decade (2010-2020), was used for the statistical analysis. The results suggest that the risk reviewer's bias, lack of close analogues and paucity of direct evidence of perspectivity, resulted in non-realistic and over/under estimation of gCOS and prospective resources. Being able to understand and quantify the risks and uncertainties, and knowing how to manage them effectively, contributes to well-founded business decisions, protects the value of projects and assets, and maximizes the value of company project portfolios. A systematic risk and peer review processes was then evolved by KUFPEC to constrain these biased subjective deviations from real objective estimations and to minimize the risk of the overestimation / underestimation of risking and hydrocarbon volume for a given prospect.