Reference dataset for rate of penetration benchmarking

In recent years, there were multiple papers published related to rate of penetration prediction using machine learning vastly outperforming analytical methods. There are models proposed reportedly achieving R2 values as high 0.996. Unfortunately, it is most often impossible independently verify these claims the input data rarely accessible others. To solve this problem, paper presents a database derived from Equinor's public Volve dataset that will serve benchmark for By providing partially processed with unambiguous testing scenarios, scientists can perform research on level playing field. This in turn both discourage publication methods tested substandard manner well promote exploration truly superior solutions. A set seven wells nearly 200–000 samples and twelve common attributes together reference results algorithms. Data relevant source code pages University Stavanger GitHub.