In recent years, operators and service companies have optimized drilling activities and reduced operational costs by accelerating the deployment of digital solutions. Among other applications, predictive data analytics are commonly used to estimate rock properties, reduce operational uncertainty, improve equipment maintenance processes, and optimize scarce human resources on specific tasks. Not surprising, then, is the number of initiatives to maximize the value of data and new machine-learning models to understand complex operations and the different levels of drilling automation around the world. The selection of technical papers for this month highlights some interesting examples implemented in diverse geographies. Sustainability appears today as a key driver for innovation. The complex process of decarbonization of our industry can be accelerated with the potential larger-scale application of emerging solutions with proven tangible impact. For instance, rig and equipment automation are a reality, with the capacity to lower emissions in drilling operations and help obtain improved and more-consistent results. Papers SPE 216249 and IPTC 22975 illustrate examples of actual benefits in different projects. The task of collecting, processing, organizing, and making sense of drilling information from multiple sources with minimum time requirements remains an opportunity for improved decision-making in project management. Paper OTC 32978 describes one case with the use of artificial intelligence for automatic classification of daily drilling reports as an enabler for better planning and accurate risk analysis. Worth emphasizing is the importance of extended reality as a growing and promising alternative with untapped potential in well construction, mainly as a resource for active learning and a collaboration tool among multidisciplinary teams. Paper SPE 212532, included in the additional recommended reading, gives an interesting overview of current applications associated with drilling operations and challenges ahead. Recommended additional reading at OnePetro: www.onepetro.org. SPE 217113 Machine-Learning Techniques for Real-Time Prediction of Essential Rock Properties While Drilling by K.W. Amadi, Australian University, et al. SPE 212532 Extended Reality and Gamification for Drilling by Crispin Chatar, SLB, et al. SPE 213043 Ensemble Machine Learning for Data-Driven Predictive Analytics of Drilling Rate of Penetration Modeling: A Case Study in a Southern Iraqi Oil Field by Dhuha T. Al-Sahlanee, BP, et al.
Read full abstract