There’s little question whether the use of AI can bring business benefits. It is also generally understood that applying AI to an organization’s emissions data should be able to bring useful insights. With businesses under pressure to reduce their emissions in the long term, there is a growing and more immediate demand to better understand what their emissions are today.
Doing so isn’t just about benchmarking for future interventions. It is also necessary in order to comply with regulations mandating emissions reporting and take advantage of business opportunities linked to emissions, such as investment credits for carbon sequestration and storage offered by the US Inflation Reduction Act.
These challenges are, at heart, data challenges, and a major roadblock for businesses seeking sustainability gains is that the data involved tends to be extensive and complex in ways that make traditional analysis and accountancy prohibitively expensive. However, there is now an opportunity to take the same AI innovations revolutionizing how businesses extract valuable insights from other data and apply it to this vital area of action.
Using AI for Emission Reduction
AI can be instrumental in identifying emissions reduction opportunities within specific areas of a business or supply chain, as well as potential investment prospects related to emissions. For AI to be effective, though, the data must be well-organized and uniform. This includes ensuring consistency in emissions data and metadata, units of measurement, calculation formulas, and component categories. Consistency must also extend to the organizational structure, including boundaries, locations, facilities, equipment, product life cycles, and the reporting framework.
Emissions data nowadays is sourced from many disparate systems, and when it isn’t standardized, the effectiveness of AI models is compromised. Although there are global GHG standards, once this data is absorbed into internal IT systems, the lack of a universal data model creates inconsistencies in naming conventions and units of measure across different business areas.
This is especially problematic for scope 3 data, which is often collected from a wide range of suppliers and business partners, each using their own formats. No matter what data integration strategies are in place, a well-structured and consistent data framework is a critical requirement for AI to deliver accurate insights on emissions across a complex organization.
Standardized Solution Across Supply Chain
The solution is to implement a standardized solution, such as the Open Footprint® Data Model, across the enterprise and its supply chain. This ensures consistency in data naming, metadata, units of measure, and the relationships between data elements.
For instance, a large multinational corporation might want to use AI to analyze the emissions profiles of various suppliers and identify which ones are successfully reducing their emissions over time. Gaining insight into supplier emissions and their progression can directly influence sourcing decisions, ultimately affecting the corporation’s scope 3 emissions. For this to be effective, emissions data from suppliers must be uniform in both data and metadata definitions. Another example involves a company evaluating its scope 1 and 2 emissions across various parts of the business, aiming to identify where capital investments can most effectively reduce emissions. Achieving this requires scope 1 and 2 emissions to be calculated using consistent data definitions, units of measure, and methodologies so that the data is comparable across different business units.
Expectations around emissions data are evolving quickly – but so are the tools we have to meet those expectations. Implementing a standardized emissions data model simplifies the collection of valuable data from across the organization and its value chain, ultimately enhancing the ability to leverage AI and unlock the business potential within emissions data.
Article contributed by Jim Hietala, VP, Sustainability and Market Development at The Open Group
