<![CDATA[The Schlumberger geoelectric method has been extensively employed in earth resource exploration due to its capability to identify variations in subsurface resistivity. However, the manual interpretation of geoelectric data inversion results is often subjective and time-consuming. This study aims to automate the lithology identification process by utilizing deep learning techniques, particularly Artificial Neural Networks (ANN), based on the inverted resistivity parameters obtained through the IPI2Win software. The Schlumberger configuration geoelectric data were obtained from survey reports provided by the Ministry of Public Works and Housing (Kementerian Pekerjaan Umum dan Perumahan Rakyat/ PUPR), which conducted geoelectric measurements in East Lampung Regency, Lampung Province, Indonesia. The ANN algorithm demonstrated an average accuracy of 90% in predicting lithology based on resistivity patterns resulting from Schlumberger inversion. Outperforming Support Vectorr Machine (SVM) (87%) and XGBoost (88%). These results confirm the initial hypothesis that ANN can effectively capture the complex relationships between resistivity values and rock types. The present study proposes an integrated approach between geophysics and machine learning with ANN algorithms for lithology prediction based on Schlumberger configuration geophysical inversion data. The present study proposes an integrated approach between geophysics and machine learning with ANN algorithms for lithology prediction based on Schlumberger configuration geophysical inversion data.]]>
Copyrights © 2025
