Integrating AI with data and experience

Artificial intelligence (AI) is transforming the energy industry, but its success depends on how well it is integrated with historical data and industry expertise, according to Wassim Ghadban, Global Senior Vice President, Digital & AI, at Kent.

Kent, which provides consulting, engineering, project management, commissioning, operations and maintenance services for the energy sector, has positioned itself as a leader in AI and digital applications for the oil, gas, and renewables sectors, Ghadban told NewsBase. The company is developing AI-powered digital solutions to enhance operational efficiency, asset performance, reduce emissions, improve safety and advance autonomous operations. It also integrates AI in order to revolutionise the way new projects are developed, from planning and design to execution.

"AI is just another technology, right? It’s going to be available to everyone and as common as Excel,” Ghadban said. “But without the appropriate data and without its appropriate integration, you will not be able to achieve the right outcome that you are looking for.”

Kent brings many decades of experience and data – “and that’s going to be the differentiator, to give you the best outcome from these AI engines.”

He stressed that AI is not a replacement for industry experience and competence, but a tool to enhance human decision-making and efficiency.

Automation

AI is rapidly gaining traction in the oil, gas, and renewable energy sectors, as companies seek cost reductions, emissions control and safety improvements. But its use in the energy industry is not new – it was first deployed in oil and gas in the 1980s, initially for seismic data analysis and later for reservoir simulation and drilling operations.

Over the last two decades, AI has expanded into predictive maintenance – helping to anticipate equipment failures before they occur – and production optimisation. However, the most significant advancements in recent years have given AI a far greater role in reshaping energy operations.

In the case of preventive maintenance, AI solutions can help significantly reduce downtime, by enabling companies to anticipate equipment failures and even automate the delivery of replacement parts from the supply chain. “You know that every minute of downtime counts in oil and gas – not only in terms of production performance but also emissions,” Ghadban said.

One of the biggest challenges facing the oil and gas industry is its ageing infrastructure, with many assets over 50 years old. This has led to fragmented data systems, inefficiencies and production constraints.

Kent’s AI-driven approach starts with retrofitting ageing assets, enhancing asset monitoring to enable operators to remotely track performance, collect data and detect issues in real time. This support can then be expanded to remote control of assets and, ultimately, their full autonomous operation.

"We are the only EPC having an autonomous operation developed in-house, where we created a virtual operator to operate any process plant in the industry,” Ghadban said. “This is how I believe Kent is taking the lead in AI application in the energy sector.”

At the core of this strategy is Kent’s advanced digital twin solution, which is designed to integrate design intelligence, project execution, operational monitoring and predictive analytics under a single framework.

"We have taken the digital twin to the next step, creating a twin for a plant before it even exists, to be able to generate scenario simulations powered by AI that will replace all the feasibility studies," Ghadban said.

This AI-powered "system of systems" allows all project stakeholders to collaborate in a seamless digital environment, ensuring that data maturity progresses from design to operations.

"This means every stakeholder engaged in the project will help really to develop the maturity level of the data as we move on from the design to operations," Ghadban said.

Workforce and maintenance support

Kent is developing an AI-powered "brain" that will assign an AI agent to every employee, including engineers, planners and controllers. Again, the success of this solution in augmenting employee capabilities and optimising project outcomes depends on its integration with Kent’s data, systems and experience, Ghadban said.

Kent is also using computer vision and AI-powered geofencing to enhance worker safety during construction and operations. AI-enabled drones and robots conduct real-time inspections, monitor compliance with safety regulations and issue alerts in case of emergency situations.

In project planning, AI can enhance human capabilities by auto-generating schedules, benchmarking projects against other similar ones and optimising execution plans, cutting down on delays and inefficiencies. There is scope to significantly reduce time spent on almost any task, according to Ghadban.

"In engineering, for instance, it may take 30-35 hours to draw a piping and instrument drawing. We are counting on these AI agents to be able to reduce this to maybe five hours," he said.

Partnership with AIQ

Kent entered into a strategic partnership last November with AIQ, a subsidiary of Abu Dhabi’s state oil company ADNOC and UAE AI developer G42, to develop AI-powered front-end engineering design (FEED) optimisation solutions. They will also work together on evaluating AI-driven innovations in autonomous operations, digital twins for asset management and health, safety, and environment applications.

Whereas Kent brings its AI & digital engineering and project management expertise and its solutions for automating plant operations to the partnership, AIQ brings its AI technology and “Robo-well” technology, which enables autonomous reservoir management.

"We formed this partnership where we count on their expertise and technologies. They developed the Robo-wells for the autonomous wells, and the plan is to integrate with our solution for the autonomous operation of process plants. Then you will be able to make the entire production autonomous," Ghadban said.

On automation of engineering design, the goal is to eventually auto-generate the entire feasibility study process – initially the creation of pre-FEED reports but later full FEED studies as well. Ghadban said the next step would be to integrate "4D and 5D simulations", combining time-based modelling, cost projections and risk assessments, alongside a digital twin platform that enables real-time insights and operational decision-making.

AI for energy, energy for AI

AI itself requires significant computing power, raising concerns about energy consumption in data centres. While AI can make both energy use and production more efficient, which supports the energy transition, the extra demand from these data centres risks increasing emissions.

Ghadban noted that recent breakthroughs are making AI models more energy-efficient, reducing their reliance on large-scale computing resources. “There are new ways to create AI solutions with less energy than before. This concept is being developed further,” he said.

AI is also being used to optimise energy use within data centres, improving power distribution and workload management, he said. And companies are shifting towards renewable energy-powered data centres, integrating solar and wind energy to lower oil and gas dependence.

“It’s AI for energy and energy for AI. Both rely on each other and it can be a win-win situation,” Ghadban said. “Because AI is addressing challenges like reducing emissions from energy and optimising production and consumption, but it needs energy to do that.

Access the full article here.