By Jim Hietala, Vice President, Sustainability and Market Development for The Open Group
The energy industry has always been a fundamentally information-driven environment.
The industry reached a point very early in its existence where substantial investment was needed for new ventures, and profits were anticipated over several decades. As such, analyzing precisely where and how to establish new operations is an existential requirement. Errors throughout the discovery process have serious repercussions that might range from restricting potential profitability to posing intolerable safety issues.
In other words, the industry has relied on big data since well before ‘big data’ became a term of art, and therefore invested heavily into it. By the time cloud computing emerged as a commodified option that businesses in any industry could access, those in energy had been working for a long time with that size of dataset using very mature methodologies.
The introduction of cloud computing represents a significant potential for the industry to approach that workflow differently. Greater speed, agility, and efficiency in analytical workflows are highly desirable in today’s data-driven industry. However, achieving these goals is challenging due to legacy methods of working that have developed over time and may conflict with one another within departments or even between businesses collaborating with one another.
The transformational power of AI
The world has become aware of a significant change in the way organizations operate due to artificial intelligence’s continuous advancements and versatility. For the energy industry, it could prove to be a vital enabler for a range of needs, from efficiently querying the vast amounts of information that operations generate, to more effectively testing actual conditions against modeled assumptions, to uncovering more optimized approaches to new exploration.
The quality of new AI applications – and, of course, the quality of all similarly data-driven tools – is dependent on the quality of the data going into them. Every industry faces this challenge differently, but for the energy industry that long history of intensive information management work, stemming from a pre-digital era and spanning a diverse range of tools developed in different places, makes it a particular hurdle.
A prerequisite to the use of AI in energy is that the data, be it exploration, production, or emissions data be well structured and consistent. Standards efforts in The Open Group OSDU Forum, Open Footprint Forum, Energistics Consortium and other standards organizations are helping drive consistent data standards enabling productive use of AI.
Industry collaboration
Working in a collaboratively structured environment means that stakeholders can share insight and expertise in a pragmatic, applicable way, and thereby modernize legacy applications in a more efficient, achievable way. Collaboration among energy stakeholders is essential for laying the groundwork for the next phase of transformation, especially as AI capabilities become increasingly operationalized in the industry.
Energy companies are increasingly establishing formal partnerships and relationships with diverse stakeholders to augment their AI capabilities. This collaborative ecosystem typically includes computing infrastructure providers, platform vendors, technology startups and academic institutions. By bringing together expertise from both within and outside the industry, energy companies can accelerate the development and adoption of AI solutions. This cross-pollination of ideas and technologies is essential for driving innovation and addressing complex challenges in the energy sector.
Addressing interoperability challenges
One of the key challenges in digital transformation is ensuring that new AI-driven approaches work together efficiently, with minimal friction in data flow. Collaboration is critical to achieving interoperability. Industry-wide collaboration on data standards and protocols can ensure that AI systems from different vendors and stakeholders can communicate seamlessly. Whilst the sharing of non-sensitive data across the industry can improve the quality and diversity of datasets used to train AI models, leading to more robust and generalizable solutions.
Addressing industry-wide challenges
Collaboration is essential for tackling broader challenges that impact the entire energy sector, for example a significant skills gap in AI and data science. Collaborative initiatives between companies, educational institutions, and governments are crucial for developing digital training courses and upskilling the existing workforce.
As AI becomes more prevalent, its energy consumption is a growing concern. Industry-wide collaboration is needed to develop and prioritize energy-efficient computing infrastructure and AI algorithms. Simultaneously addressing the risks associated with AI, such as cybersecurity threats and privacy concerns, requires a coordinated effort across the industry to develop best practices and security standards.
The energy industry is also a context where getting this right is, relatively, more critical than in most sectors. This comes back to the nature of data’s contribution to the industry’s processes. While there are situations where occasional errors, such as hallucinations in the case of generative AI, are an acceptable manageable risk, the data analysis that the energy industry relies upon is a critical input for profitability, and potentially safety, which has long-term consequences.