Digital drillship. Photo from GE.
Reducing maintenance costs is the primary focus of the oil and gas industry’s quest to get more insight from Big Data. But, can they successfully navigate remaining barriers? Karen Boman reports.
Big Data – defined as datasets of unprecedented scale – is not a new phenomenon in the oil and gas industry. For years, firms have applied analytics to these large datasets to guide decision making, particularly regarding seismic data.
While innovations continue in this arena – innovative seismic processing by BP, for example, helped push Mad Dog 2 over the line – Big Data is spreading beyond the geoscience realm, says Richard Dyson, CEO, io Consulting.
“The confluence of increased computing power, more sensor technology and the lower for longer oil price is driving the industry to look for a competitive edge through the exploitation of Big Data,” Dyson says. “The potential for insights and value to be derived from combining subsurface, subsea, facilities and operating data derived throughput the project lifecycle is huge.”
Digital Twin. Photo from DNV GL.
A major area of focus right now is using data from offshore drilling operations to reduce non-productive time (NPT) and increase drilling efficiency. Recent examples of such work include GE and Maersk Drilling’s decision to expand the scope of a 2016 pilot project, in which GE supplied SeaStream Insight software to boost Maersk Drilling’s drilling rig productivity and reduce maintenance costs by up to 20% on one asset. The software will now be installed on nine subsequent vessels, targeting 100 key equipment assets, including the top drive, drawworks, thrusters, and main engines.
Technology to gather more insight from Big Data will make the biggest impact on enhanced drilling process efficiency, says Tim Schweikert, president & CEO, GE’s Marine Solutions, if analyzed properly.
Against a backdrop of capital-intensive assets, in remote environments, with increased cost pressure and skills shortage concerns, Big Data is helping to change the way firms operate.
To reduce NPT related to machinery breakdowns has generated interest in moving from schedule-based maintenance to condition-based maintenance. In the latter, data is analyzed to determine when equipment needs maintenance work or replacement, Schweikert says.
“Therefore, maintenance actions are only implemented when needed to assure optimal reliability and save companies significant maintenance expenditures. With our customer pilot projects, we are aiming to achieve 20% reduction in maintenance expenditure through the program,” Schweikert says.
A joint development project between DNV GL and drilling contractor Transocean demonstrated the feasibility both technically and financially to ingest data from a drilling unit continuously or in batches, to aid inspections and for avoiding corrective maintenance and reducing unplanned downtime. But the processes must be automated to manage cost and increase speed, according to a presentation by the two firms at this year’s OTC in Houston (OTC-27927).
Getting more out of existing investments, extending the life of assets, and managing safety will be the main drivers behind data analytics, says Jørgen Christian Kadal, director – head of DNV GL’s analytic innovation center. Distributed computing and methods such as machine learning are enabling companies to make accurate predictions using ever larger data sets. But what will really help the oil and gas industry is cloud computing, he says. Kadal says that Big Data has actually disappeared out of strategic focus, and has been replaced by the Internet of Things (IoT) and digitalization. “Digitalization encompasses everything now, from connectivity, automation, prediction, and the processing of data, which is the Big Data part,” Kadal says.
Yet, with increased connectivity, there’s also more data: industrial IoT is bringing forth a paradigm shift in the amount of data available as sensors are collecting and communicating data from almost every possible component on a digital oil field, Dyson says. This vast increase in data not only brings about huge potential in terms of insight and optimization, it also increases the importance of data standards to ensure the data is structured to leverage the value and not present the erosion of value inherent in unstructured data.
Similarly, the concept of the digital twin relies on sensors and data gathered to accurately replicate the exact operating conditions over the lifetime of an asset such that the performance can be predicted, modeled and optimized.
For Dyson, using Big Data to create digital twins of oil and gas facilities represents one of the most exciting opportunities offered by the Big Data revolution. But, this opportunity is hampered by a lack of collaboration, he adds. Instead of working together, “there is a race to establish primacy to gain a competitive edge,” Dyson says. “This lack of collaboration hinders both innovation and actual delivery of the digital twin due to a reluctance to share data.”
The challenges in addressing Big Data are similar to those facing the introduction of any new way of working to the industry, Dyson says. First, a coherent strategy throughout the industry is needed. This is particularly pertinent for the construction of a digital twin.
“The industry needs to consider how best to develop a Big Data capability; whether that may be collaborating with Big Data companies, attracting the best talent in the industry, training our highly skilled workforce, or a combination of all these,” Dyson says.
There is good news. “As we move towards digitalization, it is producing more structured data that can be leveraged to improve business,” he says. “Digitalization is also serving to normalize data for everyone, this is helping to shift the culture from one suspicious of Big Data as the next business buzzword to one where the advantages of data are apparent and embraced. Digitalization is helping companies get more out of big data.”
Schweikert says that it is vital that the industry embraces the opportunity that transformative technologies can unlock to drive wholescale change. He noted that the oil and gas industry can look at the marine industry for lessons in looking at operations as an ecosystem, rather than dealing with individual vessels as isolated assets.
“As such, understanding that everything you do, every decision you make, is interconnected and will affect something else, is key to identifying opportunities for efficiency. We will continue to see digital, communication, and automation and control technologies being at the core of a connected marine industry,” Schweikert says.
Jørgen Christian Kadal
Yet, despite the potential for greater operational efficiencies, DNV GL anticipates only moderate investment by the oil and gas sector in Big Data technology. In a March 2016 report, DNV GL notes that, while oil and gas companies are being ambushed by start-ups and original equipment manufacturers to try new Big Data technology offerings, the oil and gas industry has been slow to adopt this technology.
This scenario remains true today, Kadal says. “One company told me that, if they wanted to, they could have meetings everyday about new analytic offerings on their existing data. They have to select the ones they want to talk to. Even though industry is slow to adopt, they’re being challenged to look at all these offerings.”
Indeed, in February this year, DNV GL launched an industry data platform, Veracity, to help the energy and maritime sectors overcome data quality issues and manage the ownership, control, sharing and use of data.
The adoption of digital technology to harness Big Data also has a downside: an increased risk of cyberattacks such as ransomware. To address the issue, oil and gas companies need to recognize the importance of information security and place it at the center of digital strategies, Dyson says.
The focus on Big Data to reduce maintenance costs is a good match of available resources and data, Kadal says. But, there’s still more for the industry to do to understand how Big Data technology could impact their operations.