Artificial Intelligence

Artificial Intelligence: 5 ways AI is disrupting Oil & Gas

20th June 2019

The Oil & Gas sector is ripe for innovation, particularly when it comes to Artificial Intelligence (AI). A recent report conducted by Markets & Markets suggested that the value of AI within the Oil and Gas industry could reach a monumental $2.85 billion by 2022 – with an astonishing compound annual growth rate (CAGR) of 13%.

Right now, the potential application of AI in Oil and Gas is broad and diverse, from process efficiencies and facilities management and safety, to forecasting, planning and surveying. And that’s just the start. We recently explored how augmented reality (AR) is already revolutionising the oil and gas sector and AR in the new enterprise. Today we’re exploring Artificial Intelligence in Oil & Gas and where we’re seeing it making the biggest waves.

ExxonMobil AI Initiative

One fantastic example of how AI is impacting the Oil and Gas industry is a recent initiative conducted by ExxonMobil. In terms of revenue, ExxonMobil are one of the top 10 largest companies in the world. They recently announced a partnership with MIT (Massachusetts Institute of Technology) to design, develop and deploy advanced robotics powered by AI. These advancements will be leveraged to generate efficiencies in ocean exploration and to optimise its natural seep detection capabilities.

The ultimate goal of the initiative is to reach a stage whereby robotic submersibles possess the same rational capabilities as the engineers who are employed to program them. This would enable robotic submersibles to perform a wide variety of tasks without any human interaction, including the ability to interpret data sets in large volumes, avoiding potential danger, and developing solutions for solving complex problems that would otherwise require the intervention of a scientist. You can dive in here to find out more.

Reservoir & Production Management

At the heart of every Oil & Gas enterprise is the oil reservoir – and it’s critical that these are managed efficiently, effectively and resourcefully to ensure their longevity. Not an easy task. Elements such as geology, production techniques and reservoir engineering activities are all involved in reservoir management operations, and they all need to work together seamlessly to optimise the reservoir lifecycle. That’s a lot of activities, a lot of people, and a whole load of budget. The good news is that by deploying AI at the heart of these operations to mine historical and real-time data, accurate reservoir characterisation – eg prediction of permeability – can be achieved, and optimal production output delivered. Testing the waters, literally and “virtually”, before the drill even touches the downhole, could save the savvy oil enterprise millions.

Oil and gas production is very different from other processes when compared to power plants, petrochemicals and refineries, in that oil and gas production presents many unknowns. Flow, well and reservoir dynamics are very difficult to fully comprehend and are constantly adapting as the reservoir is depleted and new wells are either shut in or drilled and the associated systems for managing production are altered accordingly. The data that relates to production is often difficult to interpret and often confidence is lacking in the accuracy of the measurements.

In addition, some of the sensors responsible for leveraging data can break or drift out of position over time, meaning that production teams have to continuously adapt settings. In any given scenario, there can be a vast plethora of different settings that need to be adjusted: pump speeds, pressure, temperature, gas lift allocation and choke settings to name but a few.

Depending on the scale of the reservoir, there can in some cases be thousands of different production settings to consider, with potentially billions of associated combinations of each setting. In terms of deploying AI, this means that algorithms need to be agile, adaptive and accurate at scale across a myriad of potential production combinations.

This is where ‘Artificial Intelligence As A Service’ plays a vital role in reservoir and production management. Systems are now being developed that manage data mining, machine learning and optimisation. Using highly sophisticated data harvesting algorithms, new AI systems can capture huge volumes of current raw data and compare performance with historic data sets. Using machine learning enables specific patterns to be recognised and classified in real-time and huge quantities of data can now be distilled to present meaningful information and insight. These data sets can then be analysed to develop estimation algorithms that combine advanced statistical models, known parameters, hierarchical neural networks and first principle physics. This means that the precise dynamics of the reservoir system can be fully understood and production settings can be tweaked to deliver optimal performance.

Gazprom AI Project

Gazprom, the Russian organisation who are the largest supplier of natural gas to Europe, are another great example of how AI is currently being deployed in the Oil & Gas sector. Gazprom believe that AI has the ability to generate the next productivity revolution across the industry. Gazprom have developed a partnership with Yandex (Russia’s answer to Google) to explore the commercial benefits of AI within Oil & Gas. The partnership with Yandex represents a pivotal moment for the industry as AI and Machine Learning (ML) can be deployed to manage extremely large volumes of data.

There are a number of facets associated with the Gazprom/Yandex collaboration. The collaboration will focus on drilling and well-management, modelling oil-refining strategies and the optimisation of other technical processes. The partnership will involve a review of how technologies are being used across each of these areas and prescribe best practice for new, green-field R&D initiatives. Well-management represents an enormous undertaking for companies such as Gazprom. Since commercial oil-wells started operating, over 150 years ago, drilling has been amongst the most expensive, time consuming and critical elements of oil and gas production activity. This is an area that requires significant improvement and the scope for ongoing refinement and optimisation of existing processes is vast and AI now holds the key in terms of being able to unlock additional value.

Over the past several decades, there have been significant obstacles associated with alterations in requirements for individual wells and drilling capabilities. Higher pressure levels, temperature, increased depth, directional drilling are just some of the factors underpinning these challenges. Technologies such as AI enable companies such as Gazprom to achieve ambitious commercial goals. As oil and gas assets become more sophisticated, using intelligent control systems and sophisticated operating models, the role of AI becomes increasingly important.

For other companies operating within the oil and gas sector, there’s a perception that all of the easy methods by which to optimise production performance and processes have already been implemented – the ‘quick wins’ have long since been exhausted. In order to accelerate new efficiencies, businesses will need to embrace digital transformation and innovative technologies such as Machine Learning and AI to generate new and meaningful insight from existing sets of data.

AI for Customer Services in Oil & Gas

AI also has a customer facing role to fulfil for Oil & Gas. In recent years, Shell has deployed an AI assistant to aid customer services. The two chatbots, named Emma and Ethan, provide customers with 24/7 support in relation to purchasing Shell products. The AI assistants are able to make instant recommendations based on the customer’s previous buying history and personal characteristics. The commercial impact of these AI roll-outs can be profound as intelligent chatbots can solve a wide variety of mundane tasks and enquiries, freeing up time for human employees to focus on other, more pressing areas of the business. This represents a compelling AI use case in terms of providing a hyper-personalised service, cost reduction and optimising time management for existing employees. The AI assistants possess the capability to handle hundreds of thousands of data sheets, across more than three thousand individual products.

Virtual Assistants

The usefulness of virtual AI assistants in Oil & Gas goes beyond customer support. Technologies such as AI and ML are also being utilised for safety meetings, whereby bots can be used to reference key safety details and provide actionable insights. This reduces wasted resource and duplication of effort and bots can be leveraged in some cases to make critical safety decisions in real-time. As engineering systems across the industry become increasingly complex, the use of AI will become increasingly crucial as systems adapt and become more complicated.

AI and ML are already having a serious impact in oil and gas, but the opportunities associated with discovering new use cases are vast and potentially very lucrative. There are many ways in which AI can be deployed to enhance safety and process efficiency across the entire industry which have not yet been discovered. Industry leaders in Oil & Gas have the resources and the financial muscle to take full advantage of AI and the commercial benefits it has to offer. In order to take full advantage, there are a few key considerations around senior stakeholder buy-in and resource availability.

Getting ahead in AI

In conclusion then, early adopters of AI in the Oil & Gas industry – indeed, in any industry – will likely reap the rewards of the competitive advantage that AI will provide, particularly given the pace that the technology is being adopted across the sector. The AI applications we’ve talked about here are significant in themselves … but there are many, many more as yet undiscovered, and which could have exponential impact, disrupting the industry way beyond our current thinking. The good news is that industry leaders have access to the budget to implement artificial intelligence solutions – accessing the talent to develop and implement is the challenge, and that’s where knowing what to look for and where to look will pay dividends.

If you’re a seasoned Oil & Gas innovator, or you just want to dip your toes in and test uncharted waters, Waracle can help you. If you’re ready to get the conversation started, let’s talk.

Insights like these directly to your inbox

Keep up to date with out latest thoughts
subscribe icon