Digital Oilfield: Paving the way to solving the imminent global energy crisis? Is it ready?
Exploring the potential of the Digital Oilfield in revolutionizing the oil & gas and energy industries.
With global energy consumption on the rise, the energy industry is faced with pressures now more than ever to meet the global demand. The 2017 global energy consumption was 13.511 billion tonnes oil equivalent, an annual increase of 2.2%, the largest increase in global consumption since 2013, up from 1.2% last year. Most of the global energy consumption, 57.6%, was supplied by oil & gas. [1] Clearly, oil & gas is the backbone of the energy industry and hence the backbone of our world today, fueling all technological advancements and innovations in all fields.
The question is, however, how long can we keep relying on oil & gas to keep this world going? With a world total proved reserves of 239.3 billion tonnes and an annual consumption of 7.78 billion tonnes (without considering the annual increase in consumption), our hydrocarbon resources can last for nearly just 30 more years. [1] How will we fill this gap in the future? Is there a way to optimize our oil & gas resources and prolong their lifetime to try and bridge this gap? One very promising answer is the Digital Oilfield.
[1]
Digital Oilfield
Machine learning at its core, the Digital Oilfield is a powerful combination of business process management, information technology and engineering expertise that utilizes advanced big data analytics, artificial intelligence, and the Internet of Things. Its main goal is to increase hydrocarbon exploration efficiencies and optimize oil & gas production. It can “learn” from what works in the best producing wells and apply those efficiencies to entire fields. It will predict equipment breakdown before it happens and bring about “condition-based” maintenance rather than “schedule-based” methods. Ultimately, it will produce more oil & gas for less cost. [2] The Digital Oilfield market is expected to exceed $28.53 Billion by 2023. [3]
BHGE
In July 2017, the giant General Electric completed a merger between its oil & gas business, GE Oil & Gas, and Baker Hughes, the world’s third largest oilfield service company to form Baker Hughes General Electric (BHGE), the world’s only full-stream service company. BHGE’s vision is to bring together digital solutions and technology from GE with Baker Hughes’ domain knowledge in the oilfield services sector. [4] Using GE’s Predix big data technology and Baker Hughes’ operational subject matter expertise, BHGE is at the forefront of the Digital Oilfield transformation. [5]
[5]
BHGE relies heavily on machine learning for both its process improvement and product development goals. It creates products and software to harness data from IoT connected devices and deliver machine learning predictive analytics with the goal of maximizing recovery, optimizing production, reducing nonproductive time, conducting predictive maintenance, improving safety, and enabling enterprise-wide digital transformation. Some of its major Digital Oilfield products are: Field Planning Optimization “JewelSuite”, Production & Recovery Optimization “IntelliStream”, Asset Performance Management “APM Software” and Fit for Purpose Solutions “AI Factory”.[6] In the medium/long term the company’s strategic vision is to further develop its process improvement and product development by feeding its machine learning and big data models with massive amounts of data from the digitized oilfields and to harness that pool of data to expand its footprint in the digital transformation.
Challenges & Recommendations
For BHGE to succeed in both its short term and long-term goals, it must overcome two major challenges. First, to be able to effectively link machine learning analytics with actionable and tailored solutions to clients, BHGE must focus its resources on building cross-trained teams with the operational domain subject matter expertise, the digital expertise and business knowledge. This agile, cross-functional and client centric culture must be developed for the company to achieve its goals. Second, for the big data predictive models to be effective, it must be fed with massive amounts of data which poses a big challenge in the oil & gas industry given the high level of data confidentiality between major oil & gas players. Therefore, to tackle this issue, BHGE must be focused on forming strategic partnerships with major oil & gas companies to form a hub of shared data which can be fed into the big data models.
What’s next?
- With the massive amounts of data being generated by digitized oilfields, how can BHGE manage the quality of data and make sure it’s free of noise and biases before being fed into the machine learning model to produce predictive analytics?
- In light of GE’s recent distressed financial performance and stock prices, rumors are that GE may divest its Baker Hughes business and break the BHGE conglomerate. If so, would the company have the right mix of digital expertise and operational subject matter expertise to pursue its Digital Oilfield goals? Even if BHGE was to continue to exist, is it in a financial position to be able to inject huge capital expenditures towards its goals?
Words: 799
[1] BP Statistical Review of World Energy, 67th Edition, June 2018. https://www.bp.com/content/dam/bp/en/corporate/pdf/energy-economics/statistical-review/bp-stats-review-2018-full-report.pdf. Accessed November 12th, 2018.
[2] Digital Oilfield Outlook Report. https://www.accenture.com/t20151218T203100__w__/nl-en/_acnmedia/PDF-2/Accenture-Digital-Oilfield-Outlook-JWN-October-2015.pdf. Accessed November 12th, 2018.
[3] “Digital Oilfield Market is Expected to Exceed US$ 28 Billion by 2023”. October 3rd, 2018. https://www.marketwatch.com/press-release/digital-oilfield-market-is-expected-to-exceed-us-28-billion-by-2023-2018-10-03. Accessed November 12th, 2018.
[4] Kumar, Hemanth. “GE completes acquisition of Baker Hughes for $7.4bn”. July 4th, 2017. https://www.hydrocarbons-technology.com/news/newsge-completes-acquisition-of-baker-hughes-5861820/. Accessed November 12th, 2018.
[5] DiChristopher, Tom. “GE-Baker Hughes: Some call the deal unprecedented; others see a “competitive nuisance””. 1st November 2016. https://www.cnbc.com/2016/11/01/ge-baker-hughes-merger-breaking-down-the-oilfield-services-deal.html. Accessed November 12th, 2018.
[6] BHGE’s Industrial IOT Software. https://www.bhge.com/digital. Accessed November 12th, 2018.
Great story! I am from the country that does not have a lot of natural resources. Our country spends a huge amount of money to import oil and other natural resources. That makes it vital for our country to buy stakes in foreign natural resource interests. However, such a practice exposes the country to
the risk of volatile natural resources price movement and political risks. Given the challenges and recommendations as follows you mentioned in your essay, I seriously hope technologies will solve the problems.
(1) To link machine learning analytics with actionable and tailored solutions to clients, BHGE must focus its resources on building cross-trained teams with the operational domain subject matter expertise, the digital expertise and business knowledge.
(2) For the big data predictive models to be effective, it must be fed with massive amounts of data which poses a big challenge in the oil & gas industry given the high level of data confidentiality between major oil & gas players. Therefore, to tackle this issue, BHGE must be focused on forming strategic partnerships with major oil & gas companies to form a hub of shared data which can be fed into the big data models.
That was a very interesting article! Having experience in the Oil & Gas sector, I totally understand the importance of Big Data in the future of the industry, as Owner companies seek to optimize the recovery of their existing fields, and maximize production from their new fields. I also understand the importance of of feeding predictive models with large amounts of data, and the challenges of obtaining it, given confidentiality rules that limit Owner companies from sharing their well information. To tackle this, given your point about strategic partnerships, I have recently read an article about BHGE partnering with Abu Dhabi National Oil Company (ADNOC), buying 5% equity in their drilling arm. This is a strategic partnership in my opinion, as it will enable BHGE to have access to ADNOC’s data, which can be fed into their big data model.
Interesting article! Coming from the same industry, I had exposure to BHGE’s Predix platform when they presented to my previous company and back then we had various concerns on the same. Some of it like ensuring continuous feed and the quality of the data that you have picked in the article to more intrinsic ones like transmission rate from offshore to an onshore server given the extensive dataset to concerns that whether they understand the data being fed. GE is a major supplier and OEM for offshore equipment but from the moment they sell their equipment to 5 or 10 yrs later when they might see it in their shop for bigger overhaul or repair work, they lose the insight on how their equipment is used on the field between this timeline, and what sort of maintenance philosophy is applied. To teach the ML algorithm to better predict the maintenance requirement, they not only need their own design parameters but more functional knowledge of the equipment. They not only have to extensively work on gaining this confidential data from the partners but also develop either in-house expertise or long-term partnership with users to bridge this gap. Secondly, they also need to have a strong business case on how the use of ML will gain them a competitive or commercial advantage for eg, if you go down the route of predictive maintenance, it will reduce their revenue forecast (less use of parts, services etc) so what will drive GE to use Predix to change the traditional ways of this industry?