Sequential binary classification of lithofacies from well log data and its uncertainty quantification

Understanding and identifying the composition of various lithofacies in the subsurface is essential for successful reservoir characterization in hydrocarbon exploration. However, conventional methods such as core sampling and manual well log interpretation are labor-intensive. As a result, many scientists are conducting research to utilize machine learning to study lithofacies more effectively and efficiently. However, as researchers are becoming more dependent on machine learning, uncertainty analysis of machine learning models is crucial in order to determine the reliability of the prediction results. Machine-learning algorithms that utilize ensemble methods provide an easy method for uncertainty analysis, but algorithms that do not utilize ensemble methods have difficulty in quantifying the level of uncertainty. This motivated us to introduce a method known as sequential binary classification (SBC), which helps to not only classify lithofacies but also to quantify and visualize regions of uncertainty of the machine-learning models. SBC provides a method for utilizing any classification algorithm of the user’s choice to construct an ensemble, which allows the users to readily quantify uncertainty. The proposed method utilizes the SBC algorithm to classify and quantify uncertainty from well log data obtained from the North Sea near Norway. The results show that most of the lithofacies that exist in the region of interest share similar characteristics, which results in high uncertainty among the various lithofacies, and SBC enables these high uncertainties to be visualized. We additionally demonstrate the utilization of SBC to alleviate the class imbalance issueamong the various lithofacies in the area, which is a very common problem in well log data analytics.