Transforming Data at Scale – Enabling Reliable Petrophysical Calibration through Validated and Enriched Data in Libya

Summary This paper outlines the application of an orchestrated data and AI pipeline in the classification, extraction, and visualisation of petrophysical calibration data from 3 million files. The results led to the extraction and QC of over 60,000 rows of relevant data, readily available for use in petrophysical interpretation workflows. The outlined process represents a much faster and consistent means of locating and extracting required data. Data that was then further screened to assess reliability and value for a given use case. Assessment at this level, with this number of documents, if conducted manually would not be possible or would take years. Future steps would be to improve the documentation of the transformations applied and the listing and rationale for non-extracted data.