On the conference 2nd day, on the 15th of September, Jan Kåre Igland will present “Case Studies and Results from 2.5 Years of Using Targeted Machine Learning Models to Predict Stuck Pipe Incidents”.

Several projects over the last years have aimed to utilize ML and AI technologies to improve drilling performance. They show promising results as well as challenges related to data availability, generalization, data quality and real-time enabling of the models. The objective of the presented study is to prove and demonstrate the adaptiveness to real-time data and agnosticism to well types, BHAs, mud types, lithologies or any other specific well characteristics for a set of targeted ML models predicting stuck pipe events. The models support out-of-the-box usage, which enables scalability to large numbers of wells.  The technology’s performance was successfully verified in live operations and post-drill studies on historical data on over 300 wells worldwide during the past 2.5 years, with mean recall and precision metrics of 0.986 ± 0.050 and 0.712 ± 0.181 respectively across historical test wells, and significantly reduced occurrence rates of stuck pipe.