Data & Analytics

Data Is Not Scarce, but Oil Companies Hoard It as if It Were

A growing number of oil industry leaders are saying that data sharing across the industry is needed, but change is coming slowly.

jpt-2019-05-gettyimages-1011884970.jpg

A growing number of oil industry leaders are saying that data sharing across the industry is needed, but change is coming slowly.

Recently an executive from Shell said that “unprecedented data transparency” will be needed for the oil business to adapt to rapid change. That call for action came from Mike DeWitt, vice president of operated upstream wells for Shell, during the keynote speech at the SPE/IADC International Drilling Conference, where he asked: “Is data really proprietary?”

Afterward, some delegates who heard the speech said they were struck by the call for openness in a business where that is generally not the case.

DeWitt said after the talk that while the need for change is apparent, little progress has been made. He added that Shell is supporting initiatives for more sharing.

There are multiple barriers to overcome, according to a panel at CERA­Week, an annual conference of oil and gas industry executives. Data-driven decision making represents a challenge to how engineers have historically made decisions in a business where companies shared as little as possible about wells and operations.

“Pivoting to a data-driven culture is hard enough. Now we’re trying to change the mindset from competing to partnering,” said Kentaro Kawamori, vice president and chief digital officer for Chesapeake Energy.

Digital decision analysis is a break from doing things based on experience and intuition. Companies often manage information assuming that revealing data will be giving away valuable knowledge.

“Data sharing is important,” Kawamori added. “The context of what is around it is equally important. We should get to the point of sharing more data, but we are all looking for competitive advantage.”

Finding and Producing

The urge to hoard data may be explained by a widely quoted comment that “data is the next oil.”

Both things can be valuable. But there are fundamental differences in how one extracts value from hydrocarbons in the ground and data in the cloud. Oil prices are impossible to predict, but the world’s voracious appetite for energy ensures it has value. Data’s value depends on how it is used. Having enough of the right data can help engineers find ways to reduce drilling costs, eke out more oil from the ground, or renegotiate a service contract based on productivity rather than the number of days worked.

Or it can be stored away and never looked at again, only contributing to the cost of cloud storage.

“Looking at data as oil may be like the tail wagging the dog,” said Dave Mabee, global innovation manager for ConocoPhillips. The company creates value by finding and producing oil and gas. Its people use data and analytics to help create that value.

“Thinking of data as the next oil may lead to focusing on doing things for their own sake rather than creating value,” he said.

Big oil companies have been hiring data scientists to help their staff extract more value from data, but the focus remains on helping exploration and production teams use new tools to do what oil and gas companies have always done.

For example, Shell formed a small team that has created 70 applications to allow asset teams to put together tools, like dashboards, to analyze what they are observing in the field. But that commitment is extraordinary compared with most, and they want to expand what they can do by sharing with others.

“The value of tools is not in having them; it is in using them. I am happy to give some away if we get something back,” said Theresa Baumgartner, a research engineer for Shell, who talked about the work to create a Shell app store at the SPE/IADC Drilling Conference (SPE 194100).

The desire to maximize the value of digital assets is promoting initiatives to share data and software. In the tech sector, code sharing sites such as Git Hub have become essential tools for software developers looking for modules of code they assemble.

Shell and ExxonMobil experts are among the founders of a group within SPE that is working to create a cooperative hub for data, software, and the documentation needed to make use of it all.

Persuading companies to share will be hard, but data-driven analysis creates an appetite for constantly improving analytical apps and ever more data. Hundreds of wells are often required for meaningful analysis. Even a major player with a deep database will likely want data from wells nearby owned by another company.

Data has value when it can be traded for more data. “You have some things, but you need others.” Said Paul Pastusek, a drilling mechanics advisor for ExxonMobil, who delivered a paper on the proposed drilling data and software hub at the drilling conference (SPE 194082).

His short pitch for sharing is: “Let’s combine it together and use all of it.”

Oil and Gas Sharing Initiatives

  • Microsoft and ConocoPhillips are among 97 members of the Open Subsurface Data Universe Forum, which is developing a standard data platform that will bring together exploration, development, and well data to facilitate sharing.
  • A group within SPE’s Drilling Systems Automation Technical Section is creating a cooperative repository for open-source software—code that anyone can inspect, modify, and use.
  • The newly launched UK Oil and Gas National Data Repository for petroleum-related information will offer well, geophysical, field, and infrastructure data about the offshore UK.
  • Canada’s national data repository shares volumes of E&P-associated data with organizations that produce across the country.
  • Equinor has made some data public and is working with Microsoft to share it with partners to build analytics.

Data Explosion

One link between data and oil is that companies are eager to expand their reserves of both of them.

Sensors deployed above and below ground are delivering terabytes of data a day. The industry is also using a lot more data, but it cannot keep up.

“Industrywide, companies never use 99% of the data they gather,” DeWitt said, adding that one avenue for sharing is “perhaps via academic institutions.” That would be a welcome change for the heads of petroleum programs; they have long complained about the lack of data for research and teaching.

Pastusek said that when he ­talked to professors about the data- and ­software-sharing hub: “Academics told me, ‘We have all the models we need. We want real data.”’

The appetite for data has been apparent since Equinor began offering public access to the data from a field off Norway called Volve. It produced 63 million bbl of oil from 2008 to 2016. Among the partners on that field is ExxonMobil.

The data release spawned the development of software to mine it. The data also was used by other companies to develop products ranging from fast reservoir imaging software to a tool for storing well data in one place.

Recently, the UK Oil and Gas National Data Repository (NDR) began offering well, geophysical, field, and infrastructure data about the offshore UK. Some data have been given to the group within the SPE Drilling Systems Automation Technical Section to create a hub for sharing. That is a small start compared with the ambitious goals of the hub’s founders.

Their technical paper summarizing their plan lists the many topics of interest and the documentation required to make sense of it all. It also explains the need, and touches on the legal issues related to make an open-source software and data hub a reality.

Tough Problem

Pastusek acknowledges that the open-source group is embarking on a difficult, long-term effort. “You know we (oil companies) don’t share data well,” Pastusek said. “It is hard to get that approval—and can take a lot of time to do so.” The difficulties are magnified when partners are involved because each must approve the release of software or data from the project.

DeWitt pointed out that the change is possible when there is a sufficient reward. Companies often partner on big projects to share the cost and risk.

In the early 1960s, Shell’s development of the first floating drilling rig, the Blue Water I, created a daunting regulatory obstacle. “They (regulators) could not grant leases because no one else could bid on them,” DeWitt said.

The solution was to share the technology it developed so other oil companies could compete for leases. Shell did not profit from its ingenuity, but that allowed Shell, and the global energy supply, to reap a far larger payoff from deepwater discoveries.

Now the pressure to share is coming from the shale oil business, where the industry is betting that better data and analytics will deliver the big productivity increases needed to satisfy investors.

Kawamori said sharing subsurface data could help improve shale well productivity at a time when the industry badly needs to increase its return on the many wells drilled to develop shale plays.

“Data is emerging showing that the amount of cash this industry has burned is not sustainable. We all want to be capital positive. If we all knew where we were drilling, what we were drilling, and how to do it more efficiently and profitably, we all would win.”

Information related to safety or the environment has been an area where there has been sharing, and DeWitt said openness on ways to improve operations should be another.  

He pointed out that in mature basins, companies have long-established developments where benefits of learning from other operators far exceed any corporate intelligence shared. An initial area for sharing would be in narrowly targeted performance indicators that he called micro-KPIs.

One area where Shell is sharing is with a select group of service providers with contracts based on performance rather than day rates. The data is needed by service providers when negotiating deals with compensation based on performance to know if the targets are realistic up front, and to measure how those providers performed versus performance benchmarks.

“The industry will not be sustainable if we hide behind contracts and clauses as we have in the past,” DeWitt said.

The benefits of digital tools will depend on changes in the mindsets of those in the organizations.

“Cheap access to power computing and algorithms are available. What we need is a cultural shift,” said Craig Hayman, CEO of British information-technology company Aveva.

“Suppliers will compete with each other to provide the best applications and services using data from an open, standards-based ecosystem. Customers then will be able to select best-of-breed services and further differentiate themselves by developing proprietary applications. The competitive advantage will come not from the data itself, but from how it is used,” he said.