Why Oil And Gas Majors Are Spending Billions Of Dollars On AI?
The oil and gas sector is increasingly deploying artificial intelligence (AI) when drilling wells, ... [+] finding productive fields, optimizing processes, reducing downtime and improving health and safety.
We are unquestionably in the midst of a global public debate about artificial intelligence (AI) and machine learning. The discussion often involves two alternate and somewhat extreme realities – a dystopian vision wherein advanced digital minds and machines would have grave consequences for humanity, and a utopian one that sees AI driving progress, efficiencies and economic growth in a futuristic setting. What ultimately unfolds will likely be a more balanced and process oriented deployment of AI.
As the commercial and industrial world makes its own mind up whilst grappling with the political classes, one segment of the global economy is clearly voting in favour of process oriented AI – the oil and gas sector. Quite used to cyclical volatility and economic downturns impacting their fortunes, hydrocarbon extractors are finding an ally in AI to facilitate operational predictability as well as assistance in meeting their carbon emissions targets.
Observed and anecdotal evidence suggests they are on the cusp of some very profound changes. For instance, all of the top 20 global oil and gas producers, be they state-owned entities or public-listed ones, have a clear AI strategy for their upstream (i.e., exploration and production), downstream (i.e., processing and refining) and, where applicable, midstream (i.e., pipeline and logistics) businesses.
Additionally, a recent EY survey indicated that more than 92% of oil and gas companies around the world are "either currently investing in AI or plan to in the next two years." Aware of their clients' needs, technology-industrial software vendors as well as major oilfield services companies (e.g., Baker Hughes and Halliburton) routinely offer AI solutions for more efficient technology-enabled operations.
Talk of predictive maintenance, advanced analytics, machine learning algorithms and data driven process optimization is no longer the preserve of some tech jamboree in the Silicon Valley. Increasingly, oil and gas sector events are allocating huge chunks of their discussions to AI.
Offering a case in point is Gastech, a key industry fixture that's been running for over 50 years. The event's upcoming round in September 2023 has an entire conference stream dedicated to the subject, alongside mainstream industry discourse on natural gas and hydrogen markets. Even the recently concluded Organization of Petroleum Exporting Countries (OPEC) International Seminar agenda deviated from its traditional geopolitical and market discussions to technological investments.
As for the industry's capital expenditure (capex), several forecasts are floating around the market under the digital transformation umbrella. But more specifically on AI in oil and gas, Mordor Intelligence projects it to hit $2.38 billion by end-2023, rising to $4.21 billion by end-2028; a CAGR of 12.09% during the forecast period (2023-2028).
There are obvious use cases for AI within the sector, also seen elsewhere in the global industrial complex. These include AI-led raw materials procurement, inventory, logistics, operational decision making, back-office management, and of course, AI-premised cybersecurity, given the times we live in and the strategic importance of energy infrastructure.
Optimizing hydrocarbon production by using AI and machine learning.
Beyond these are several sector-specific investment pathways. These include capex allocation towards AI-led geological assessment, seismic data and surface analysis in hydrocarbon prospection. We are not only talking of computation prowess for data analysis but AI-powered robots scouring drilling sites, saving time, enhancing efficiencies, improving margins and lowering the carbon footprint for core operations.
Another hot AI investment focus area is reducing downtime at gas and oil wells, platforms, pipelines, refineries and other assets, as unplanned stoppages cost the industry millions of dollars. Leveraging AI, companies are now using predictive analytics to anticipate anything from the timing and duration of refinery shutdowns and maintenance, to pipeline blockages and well collapses. To varying operational scales, AI algorithms now routinely analyze incoming data for anomalies and process inconsistencies, ultimately raising red flags in the monitored equipment.
The holy grail of management is the deployment of "digital twins", or virtual cloud-based replicas of an operational piece of upstream, midstream or downstream process or equipment. It is underpinned by first principles schematics / models / workflow charts, machine learning and process response software. Combining one or more of these aspects, digital twins generate simulations that can anticipate operational needs and prevent mishaps.
According to a whitepaper published by Honeywell, a 20-year veteran proponent of the technology, digital twins can be deployed for assets (e.g., wells, electrical submersible pumps, compressors), or processes (e.g., corrosion monitoring, oil / gas lift optimization, emissions anticipation and tracking). And AI is getting bigger by the minute for workplace health and safety management with the deployment of virtual and augmented reality (VR / AR) training, onsite advanced robotics and data drones.
Understandably, capex allocation to these disruptive themes is driven by necessity and costs. However, there is another key motivation – management of the energy sustainability trilemma (sustainability, security and affordability) – and ensuring that there is an equal focus on all three aspects.
Industry pragmatists are betting on AI solutions to keep production costs down that can be passed on to consumers, enable more efficient ways of securing and maximizing resources, and help with meeting emissions targets over time. As the billions being on invested on AI come into sharp focus, potential gains (or otherwise) from it would soon be under the microscope too.
Finding an ally in AI