Enhancing Proactive Well Management: Schlumberger's Use of Snorkel Flow

• Analytics & Modeling - Machine Learning
• Sensors - Liquid Detection Sensors

• Education
• Mining

• Maintenance
• Product Research & Development

• Time Sensitive Networking
• Virtual Training

• Data Science Services
• Training

Schlumberger, a leading provider of technology and services for the energy industry, faced a significant challenge in extracting crucial information from a vast array of daily reports. These reports, ranging from daily drilling reports to well maintenance logs, each had their unique structure and format, making it difficult for Schlumberger’s team to quickly extract the necessary information. The team attempted to automate the information extraction using Named Entity Recognition (NER), but off-the-shelf ML models failed to identify the scientific terms related to the Exploration and Production (E&P) industry. Creating a domain-specific training dataset was time-consuming and not scalable, taking anywhere from 1-3 hours per document. The team needed to identify 18 different industry-specific entities and automatically associate data with these entities. However, the rich information was buried within tabular and raw text in PDFs with varied formatting across reports from different companies. There was also poor collaboration between domain experts and data scientists, with cumbersome file sharing and ad-hoc meetings.

Schlumberger is the world’s leading provider of technology and services for the energy industry, operating in over 120 countries. The company provides well maintenance and analytics services to the world’s biggest oil companies. Schlumberger believes that large-scale data analysis and artificial intelligence/machine learning will help them remain a leader in the market. The Software Technology Innovation Center (STIC), within the 85,000-person industry leader, is dedicated to using new AI/ML applications to support the company’s mission to improve the performance and sustainability for the global energy industry. They aim to streamline information extraction from critical field data that underpins Schlumberger’s efforts to do a large-scale analysis of business operations and deliver data-driven insights into their performance.

Schlumberger turned to Snorkel Flow to build an AI application that could automatically extract key scientific data from geological and field data reports. This solution was developed in just three days and was used to guide recommendations for better well management across multiple clients. The AI application was built using a data-centric artificial intelligence (AI) development lifecycle accelerated by programmatic labeling. After a few rounds of rapid iteration using Snorkel Flow’s model-guided error analysis and programmatic labeling, the team improved their F1 score to 91.4%. The AI application reduced the processing time of reports from 1 to 3 hours per report to just a few seconds. It was able to extract several different entities from unstructured data, including well maintenance activity description (textual), time of activity (numerical), and more. The solution also overcame the challenge of non-standard reporting formats, successfully identifying entities across 15 different document structures.

• The AI application built with Snorkel Flow not only significantly reduced the processing time of reports, but also improved the accuracy of information extraction. The solution was able to generalize to a variety of document structures, including unseen PDF and tabular formats, thereby overcoming the challenge of non-standard reporting formats. This led to improved collaboration between domain experts and data scientists across labeling, troubleshooting, and iteration. The solution also enabled auto-labeling by capturing labeling expertise as labeling functions and applying them intelligently en-masse. The successful development of this AI-enhanced tool has established a repeatable data-centric AI development lifecycle as a foundation for the future of data science development at Schlumberger.

• Built a highly-performant ML application in less than 3 days
• Improved F1 score to 91.4% after a few rounds of rapid iteration
• Reduced the processing time of reports from 1 to 3 hours per report to just a few seconds