Q&A: AI and the energy transition

January 2024  |  SPECIAL REPORT: ENERGY & UTILITIES

Financier Worldwide Magazine

January 2024 Issue

FW discusses AI and the energy transition with Riz Sahib at BDO.

FW: How would you characterise the scale, cost and complexity of decarbonising the global energy infrastructure system?

Sahib: Decarbonisation investments are separated into three tiers. In the first tier are the relatively lower cost ‘quick win’ opportunities that reduce emissions and investment costs quickly and thus help generate more significant returns. An example is energy efficiency. The second tier, transition investments, require more capital. Industries like manufacturing, natural resources and infrastructure can invest in technologies that optimise operations to reduce emissions and manage operating costs. Government incentives encourage these investments, and early adopters attract investors who are proactively managing climate risk. The third tier of investments are transformations, and they require both significant capital and an ecosystem of producers and buyers. A good example is the development of the clean hydrogen ecosystems that require an investment in infrastructure from electricity, oil and gas. These industries will also need to start investing in technologies to leverage green hydrogen.

FW: Could you provide an overview of how artificial intelligence (AI) is aiding in the energy transition and efforts to reach net zero by 2050?

Sahib: Artificial intelligence (AI) makes solving complex problems and optimising multiple variables infinitely easier. AI solutions can support all stages of the energy transition, including quick wins, transition and transformation. We see AI in smart building solutions to reduce energy use and integrating onsite solar, electric vehicle (EV) charging and battery storage. By optimising logistics and production processes, AI also supports efficient supply chain operations. As energy infrastructure needs evolve, AI will help emerging technologies better predict grid demand and model the best ways to reliably meet that demand, while also balancing costs and emissions.

FW: What types of AI technologies are typically being deployed toward this goal? To what extent are digital solutions complementing physical innovations?

Sahib: The most common use of AI in the energy transition is in forecasting, including of energy demand and optimisation. It helps us identify the best mix while balancing reliability, cost and sustainability. Manufacturers are also applying this type of strategy to battery storage, wind, solar and hydrogen electrolysers. AI-powered forecasts and optimisation also streamline manufacturing processes like production and waste and emission reduction. For logistics and distribution, AI technology’s ability to predict demand for goods can support route optimisation, load management for transportation, and warehouse management to support just-in-time delivery.

“AI can create tremendous value and often has an internal rate of return of 40 percent or more, but upfront costs can be high.”
— Riz Sahib

FW: Are there any specific examples that demonstrate how AI applications have been used to significantly improve energy efficiencies and cost savings?

Sahib: The benefits of AI in managing energy efficiency shine with building management. For instance, a technology company like Grid Rabbit that specialises in smart buildings has a hospitality solution which leverages smart sensors to track hotel traffic in rooms. Integrating sensors with booking systems can help hotels reduce energy use by 20 percent. The platform balances guests’ comfort, energy consumption and emission reductions. It uses AI to predict room occupancy, uses weather forecasts to understand how external temperatures will impact humidity and temperature in the hotel, and develops a strategy to manage the environment in individual rooms. In more advanced use cases, the platform has integrated solar, battery and EV charging to manage energy demand to limit peak demand charges by balance load and local energy production. This solution has reduced energy costs by 35 percent, with a payback period of two years.

FW: What are the key data and digitalisation challenges involved with using AI for the energy transition? How much is dependent on data quality?

Sahib: In most AI use cases, accessing data and managing data quality is a common challenge. Most data is in operations, finance and other platforms that are not designed to easily extract information or run complex AI. Doing so on the cloud can help manage demand for computing capacity to run the AI models. To maximise AI, high quality data is critical, both for forecast and optimisation strategies. Learning how to manage missing data and identify primary, backup and tertiary data sources to manage error handling and data quality issues are critical to using AI effectively.

FW: What advice would you offer to energy companies seeking to deploy AI in their operations? What considerations need to be made to manage risks and maximise benefits?

Sahib: Organisations that are deploying AI should first identify the problems they aim to address and the outcomes they want to achieve. Then, they should clearly define value drivers like revenue and cost savings. AI can create tremendous value and often has an internal rate of return of 40 percent or more, but upfront costs can be high. Setting expectations for what success looks like at the outset to avoid scope creep is imperative. Another challenge is adoption. Programme managers need to listen to end users and make sure they are championing the programme. End users create the value, so they must be comfortable with the information that AI solutions are providing. Many great ideas fail because the end user has not fully bought into the solution. Accountability for plant operations and financials still sits with individuals, so trust is paramount.

FW: Looking ahead, what are your predictions for the uptake and application of AI solutions as a potential game changer for the energy industry and its transition efforts?

Sahib: We expect the uptake of AI solutions in the energy transition to parallel the story of how application development gradually became more accessible. While programmers initially faced barriers to entry, they can now develop applications using dedicated platforms like PowerApps. Similarly, over time, AI will become more accessible to more users. In parallel, landmark advances in quantum computing will create new applications of AI to address more complex problems like grid planning and optimisation, which computing capacity currently limits. With quantum computing, AI can be fully implemented to help drive better decision making and resource allocation.

 

Riz Sahib leads the climate mitigation practice for BDO in North America. He has been working in the energy and sustainability space for the past 15 years. He helps organisations develop and implement sustainability and decarbonisation strategies, leveraging process improvement to reduce waste, using digital transformation to optimise asset performance, and enhancing renewable energy and fuel procurement strategies. He can help to identify opportunities to invest in infrastructure like water reclamation systems, electrification of processes, and microgrids to reduce emissions, waste and operating costs. He can be contacted on +1 (508) 314 9513 or by email: rsahib@bdo.com.

© Financier Worldwide

THE RESPONDENT

Riz Sahib

BDO

©2001-2024 Financier Worldwide Ltd. All rights reserved. Any statements expressed on this website are understood to be general opinions and should not be relied upon as legal, financial or any other form of professional advice. Opinions expressed do not necessarily represent the views of the authors’ current or previous employers, or clients. The publisher, authors and authors' firms are not responsible for any loss third parties may suffer in connection with information or materials presented on this website, or use of any such information or materials by any third parties.