Data Analytics

Digital transformation supported by data analytics and artificial intelligence can be the differentiator that allows the upstream industry to persist in the next 100 years, providing a unique foundation for innovation to enhance general public awareness and life quality.


Until a decade ago, the practice of data analytics and artificial intelligence in upstream was sporadic, and that was only possible through research organizations and individual effort. Nowadays, we see an increasing trend of image-classification applications for reducing subsurface studies from months to days and time-series analytics for recommending actions and preventing failures in real time.

Despite all efforts, the hydrocarbon industry continues to be perceived by many as the black sheep of all the energy supplies; its sustainability is jeopardized by environmental concerns and the net unit cost. In addition, operating companies are charged by shareholders to increase profitability continuously, leading to longer-than-usual working periods and less-safe working sites.

Digital transformation supported by data analytics and artificial intelligence can be the differentiator that allows the upstream industry to persist in the next 100 years, providing a unique foundation for innovation to enhance general public awareness and life quality.

Because automated machines yield more results per unit time and less unit cost, genuine concerns have arisen about artificial intelligence reducing the number of jobs or making some positions obsolete. Numerous cases exist in which automated oilfield operations have delivered safer workplaces with fewer human-intensive decision-making processes. These operations and processes run 24/7 with very high availability, reducing people-power requirements by 80% or more.

Here, three examples are highlighted that relate how digitization, analytics, and artificial intelligence are transforming how geoscientists will work in the future. Extracting more information from seismic and log data and deriving reservoir states without simulating the porous media are becoming common feats, thanks to high-performance hardware and data analytic applications.

In dealing with the digital world, professionals need to carry a set of particular skills, including computing upgrades; system maintenance; and data manipulation, including from exploratory data analysis and programming of exception-based surveillance rules.

Some places in our industry, how­ever, exist where machines cannot replace the human touch. The future petroleum engineer, released from traditional mundane tasks, would need to focus mostly on creative work, including, for example, the creation of innovative porous-media recovery mechanisms, ground-breaking business models for hydrocarbons in society, new uses of environmentally friendly materials, and cost-effective facility life-extension options.

The future hydrocarbon industry, enabled by digitization, analytics, and artificial intelligence, will demand more creative and innovative mindsets and skills able to ingest analyses from multiple domains; this, in turn, could lead to a safer and more profitable industry with fewer working hours and improved work/life balance.

This Month's Technical Papers

Data-Driven Analytics Provide Novel Approach to Performance Diagnosis

Artificial Intelligence Improves Seismic-Image Reconstruction

Data-Driven Tool Uses Amplitude-Based Statistics To Identify Seismic Fractures

Recommended Additional Reading

SPE 193080 Hybrid Artificial Intelligence Techniques for Automatic Simulation Models Matching With Field Data by Marco Giuliani, Eni, et al.

OTC 29415 Validating Drilling-States Classifiers With Suboptimal Data Sets by Luis R. Pereira, Transocean, et al.

SPE 191643 Inferring Well Connectivity in Waterfloods Using Novel Signal-Processing Techniques by Y. Wang, University of Oklahoma, et al.

Luigi Saputelli, SPE, is a senior reservoir engineering adviser with ADNOC. During the past 25 years, he has held various positions as reservoir engineer, drilling engineer, and production engineer. Saputelli previously worked for 3 years with Hess Corporation, for 5 years with Halliburton, and for 11 years with Petróleos de Venezuela. He has worked in multiple global assignments in countries such as Venezuela, Argentina, Brazil, Colombia, the UK, Thailand, Malaysia, Indonesia, Australia, Saudi Arabia, Kuwait, the UAE, and the US. Saputelli is an industry-recognized researcher, invited lecturer, and SPE liaison and committee member. He is a founding member of the SPE Petroleum Data-Driven Analytics technical section and recipient of the 2015 SPE International Production and Operations Award. Saputelli has authored or coauthored more than 70 technical publications in the areas of digital oil fields, reservoir management, reservoir engineering, real-time optimization, and production operations. He holds a BS degree in electronic engineering from Universidad Simon Bolívar, an MS degree in petroleum engineering from Imperial College London, and a PhD degree in chemical engineering from the University of Houston. Saputelli serves on the JPT Editorial Committee, the SPE Production and Operations Advisory Committee, and the Reservoir Description and Dynamics Digital Oil Field subcommittee. He has served as a reviewer for SPE Journal and SPE Reservoir Evaluation & Engineering and as an associate editor for SPE Economics & Management. Saputelli also serves as managing partner at Frontender, a petroleum engineering services firm based in Houston. He can be reached at