AiLAB: An In-house Developed Codeless Platform for Machine Learning Modeling

As digital transformation (DT) is unleashing unprecedented potentials to maximize the utility of subsurface data, the machine learning (ML) technology has become ubiquitous, enabling unprecedented opportunities in the energy industry. It is now a common assumption that coding skills is a major requirement to be part of the ML-driven DT era. The reality in the industry dictates that everybody may not have the interest, time, and space to acquire the coding skill. To prove this point, this paper presents AiLAB, a codeless platform developed in-house to help domain experts and ML enthusiasts to be able to benefit from the capabilities of the ML technology without any limitations. We demonstrate the capabilities of this platform by using it to execute a project that explores the feasibility of predicting reservoir rock porosity from advanced mud gas data. We followed the ML standard implementation procedures of feature selection, data cleaning, data analysis to confirm the expected input-target correlation, data splitting, ML modeling, model validation, model selection, and deployment. The validation results showed that the random forest model outperformed the multilinear regression, artificial neural networks, and decision trees using the common metrics including correlation coefficient, mean squared error, and mean absolute percentage error.