Select Country
Follow Us

Making Fracking More Efficient With Prescriptive Analytics

It’s difficult to make fracking more efficient and safer at the same time, but we can hedge our bets using the advanced data technologies of “prescriptive analytics.”

There’s a lot of data moving through a fracking operation. Depending on the shale play and the operator, the drilling phase can last from six to thirty days, the fracking-completion phase can last from three to eight days and production phase can last for 30 years. And all that data comes in the form of different datasets. The big question is how to make sense of them together?

To get more oil and gas from a well for a longer period of time and a lower environmental footprint, it’s crucial to extract actionable information after combining all these datasets. As the industry gets into the “shale manufacturing” phase, and as exploration and production costs rise, an operator’s ability to improve production performance and safety is instrumental for success in the highly competitive marketplace.

As oil prices start dipping gradually over the next couple of years, the ability to make accurate investment-drilling-fracking-completion decisions will become a must-have competency. Enter big data analytics.

The first phase of big data analytics is “descriptive analytics,” focused on past and present. It answers: What happened? Why? What is happening? Why?

The second phase is “predictive analytics,” focused on the future. It answers: What will happen?

The third phase, the “final frontier,” of big data analytics is “prescriptive analytics” and is focused on improving the future. It answers: What will happen? When? Why? How can we improve this predicted future without disrupting other priorities?


Better for the Environment
Today, most oil companies have drilled only a fraction of their unconventional acreage. They’ve learned valuable lessons—and collected critical data— from these drilled-fracked-completed-producing wells. Now the question becomes: how can an operator unearth and subsequently use the insights lurking in all these datasets to improve production from its remaining acreage positions?

Development of unconventional shale plays involved the drilling of tens of thousands of wells over hundreds of thousands of acres. By using prescriptive analytics, we can minimize the number of wells and pads needed to efficiently drain the reserves. This can be accomplished by analyzing information from geophysical and well log data, as well as the thousands of treatment stages to truly understand which combination of variables has the biggest impact on ultimate production.

If we can make each well treatment more effective and place each well in a suitable location, we will greatly reduce the overall footprint at the surface and minimize the amount of water and chemical usage downhole.

Drilling:A properly constructed well bore prevents blowouts and protects fresh water sands during hydraulic fracturing operations. By analyzing 3-D seismic data, real time drilling logs, pressure while drilling, weight on bit, drag, torque changes, cement bond logs, and many more data measurements, one can optimize the well design sooner in the fields development, and protect sensitive fresh water from the very inception. Also, from a video standpoint, training prescriptive analytics algorithms to recognize and predict unsafe behavior and sending a warning—with the suggestion for a preemptive measure—to the on duty safety supervisor can prevent accidents and save lives.

Fracking-Completion:To manage dangerous high pressure, the key is in integrating video surveillance with data from pumping and wellhead equipment to preempt an uncontrolled or catastrophic release of frack water and hydrocarbons.

Production: The goal is to minimize oil spills and unintentional gas flaring. Oil and gas companies collect a wealth of real time data—sometimes, every few seconds—on each piece of equipment. The data currently just alerts someone if a failure occurs, but prescriptive analytics could alert an operator before a failure occurs with a recommendation for an appropriate preemptive measure.

Better for the Bottom Line
Emerging technologies like fiber optic sensing combined with existing diagnostic tools like microseismic are now showing us that the current fracturing process isn’t optimized. Many times only a small fraction of the reservoir is stimulated with each fracture treatment. By combining all the available datasets with prescriptive analytics, we can make better decisions looking ahead rather than review the terabytes-petabytes of information after the fact and try to make incremental improvements on the next well.

Prescriptive analytics can offer data-driven answers to the mission critical questions:

  • Prediction: Where to drill? Why?
  • Prediction: Where to frack? Why?
  • Prescription: How to complete each well for maximum production?

The good news is the pioneers in the energy business are starting to pay attention to prescriptive analytics. Google-like technologies are finally arriving to improve some of the most expensive, in more ways than one, decisions in the planet.

Atanu Basu is the CEO of Ayata, a software company that invented and refined Prescriptive Analytics over 10 years.

Making Fracking More Efficient With Prescriptive Analytics was originally published on Ideas Lab

Subscribe to our GE Brief