AI/ML is seen as a key enabler of digital transformation by the E&P companies. But what enables AI/ML? The following article is the first in a 4-part blog series, which explains four success factors that can improve the value and impact of AI/ML initiatives in the E&P industry.
These days, more E&P firms see AI/ML as a core capability for their business, resulting in a surge in AI/ML initiatives, and buzz that refuses to die. But it’s also true that the industry is still far from realizing the true potential of the technology. The last decade was a mixed bag for AI/ML. Many of the projects struggled at the pilot stage and very few survived the operational deployment stage.
The most typical issues weighing down AI/ML initiatives include: misguided initiative selection, unrealistic expectations, limited user adoption, and inadequate planning. These issues may not have received much attention or sound bites in the industry literature or conferences, so far, but it could well be chalked up to a mostly blinkered and hype-tinted view of AI/ML as just a technology cure-all.
From a Landmark point of view, the more digitally advanced E&P firms - the ones already ahead on AI/ML experience curve - are thinking differently. Their executives, along with the industry digital experts, are stressing the need to embrace a more thorough approach while planning and executing AI/ML initiatives, addressing the specific organizational, operational, and data issues.
To that end, the following four success enablers are seen to particularly resonate with the serious-minded E&P firms.
- Talent Transformation to develop AI/ML-savvy workforce
- Trust-Based Modeling through integration of AI/ML into petrotechnical applications to create user trust in the algorithms
- Open and scalable architecture to manage operationalization complexity
- Governance for model management for long term sustainability

Fig: Four Enablers of Success with AI/ML
This blog concentrates on the role of Talent Transformation.
Talent Transformation: Bringing Data Scientists and Domain Experts Closer Together
Developing and sustaining AI/ML systems demands a symbiotic collaboration between data scientists and domain experts. Domain experts should be able to spot the problems that are amenable to AI/ML application. Data scientists should appreciate the context and constraints of the petroleum sciences within which their algorithms need to work.
Here, E&P companies face challenges on two fronts: First, there are persistent gaps in understanding how AI/ML works, even at a conceptual level. In our interactions with the E&P workforce, we often come across a mythical belief in the power of technology. This leads to unrealistically high expectations and end users potentially choosing the wrong problems to solve with AI/ML.
Second is the general difficulty in finding good data scientists, let alone the ones who have an E&P industry background. It’s not only because of growing competition for AI/ML skills across the business sectors. Additional complicating factors include shortage of easily accessible data sciences courses for the E&P workforce, and the industry’s diminishing popularity as an employer of choice.
This dual challenge is not lost on the E&P companies. Some senior executives we have spoken to expressed the urgency to grow their own analytics talent pool. A few companies have already increased their data sciences competency development efforts by tapping into analytically oriented minds in their workforce.
Many others, however, lack the willingness or wherewithal to do so and become overly dependent on external help. This also makes them tentative and cautiously slow in their approach, spending significant time on precautionary due diligences on their suppliers. AI/ML initiatives are seen to drag on in such cases.
How Is Halliburton Helping E&P Companies Transform Their Talent?
Halliburton works with E&P companies to develop their analytics talent in order to accelerate their AI/ML initiatives and collaborate confidently with the external suppliers. We enable talent transformation through our industry- leading data sciences competency development program that includes field-proven approach and methodology, and content tailored to the oil and gas domain.

Fig: A Snapshot of Halliburton’s Data Sciences Competency Building Track Record
Comprising practical workshops and boot camps with hands-on exercises, our training grounds users in the fundamentals of AI/ML, its real world applications, and different model-building techniques. Our courses, contextualized for asset lifecycle challenges, use publicly available or client data sets.
All our training is provided using widely available open source technology and the OpenEarth® Community. This helps us maintain our technology neutrality and avoid biasing our curriculum to favor a specific vendor technology.
The courses we deliver are tailored for multiple audiences (from technical users to executives), different disciplines (e.g., G&G, Reservoir, and Drilling), and varying skill sets (from beginner to advanced).

Fig: Sample Courses Taught by Halliburton Data Scientists
Our unbiased and practical training approach helps E&P users build well-rounded knowledge of the inner workings of AI/ML and get more value out of their datasets. More importantly, users also learn to appreciate the limitations of AI/ML and how it reacts to evolving operating and data environments. This understanding not only improves users’ engagement with the technology but also drives its clear-eyed application to business problems.
Benefits extend well beyond that, though. A well-informed organization is more likely to select the right AI/ML projects and avoid hype-induced moonshots that tend to be a drain on the organization’s resources, delivering very little value and plenty of disappointment.
Summary
Competency development is just one element of talent transformation. E&P companies must work towards creating an organization environment that supports and incentivizes newbie petroleum engineers and geoscientists to develop their data sciences skills and apply those to their work.
Asset managers and discipline heads must work with their human resources department to define a talent retention plan including formal career paths for the citizen data scientists. Without that, they risk losing their hard-earned talent to a highly competitive market. Ultimately, we believe success in using AI/ML to transform the business will invariably come to those who can successfully build and retain the right digital talent. E&P companies looking to reach that position of advantage must start investing in shaping their workforce so that they can help drive the digital transformation of their organization.
Click here to learn more about what we’re doing in Talent Transformation in the oil & gas industry.