Abstract
<jats:title>Abstract</jats:title> <jats:p>Reservoir static pressure measurements are fundamental throughout the oilfield lifecycle. From exploration to development, wireline formation testing tools deliver accurate reservoir pressure and dynamic characterization of formation and fluid mapping. Despite their importance for petrophysical evaluation and reservoir understanding, these operations are often time-consuming, impacting in cost-efficiency in deepwater development projects. This paper proposes a workflow using automation and AI to increase efficiency and consistency in acquiring these measurements, while ensuring reliability and operational safety.</jats:p> <jats:p>Following a continuous improvement approach and looking for optimizing operations in a Brazilian presalt deepwater development campaign, we applied a methodology for automating the acquisition of wireline formation testing tools coupled with Machine Learning models for pretest labeling within 60 seconds. The objective was not only to enhance acquisition efficiency, but also to improve data quality by reducing human-related errors and ensuring a standardized workflow, while maintaining safe options for human intervention. Within the same presalt field, this methodology was deployed in three wells, and results were systematically benchmarked against previous wells using the same wireline formation testing tool.</jats:p> <jats:p>In the wells where the methodology was applied, the average time per pretest station was reduced by approximately 70%, resulting in savings of up to 7.4 hours of rig time per descent, considering the number of pretest points. These results demonstrate that automation and machine learning can significantly improve efficiency and consistency in reservoir pressure acquisition, while maintaining data quality and operational reliability, generating measurable rig cost savings in deepwater operations. Moreover, by shortening the operation duration, we significantly reduced the personnel exposure in the rig red zone during wireline operations, where most accidents by dropped objects are typically reported.</jats:p> <jats:p>This methodology demonstrates the successful deployment of automation and machine learning in wireline formation testing for Brazilian presalt deepwater wells, delivering significant efficiency gains, standardized acquisition, and improved data quality, with strong potential for scalable application in other offshore developments.</jats:p>