The climate crisis threatens the supply chains. Manufacturers hope that AI can help
Abi Ghadj, an assistant professor of supply chain management at the University of Karnfield in the UK, says there was a “general type of negligence” in terms of climate resilience, though it starts to change.
However, it is very difficult for smaller companies to create a precise understanding of a supply chain. Who does their suppliers provide? Which key raw material is being exposed to deficiency? Beatriz Royo, Assistant Professor of the Mit-Zaragoza program in Spain, says tracking such details requires long-term commitment and investment.
Given this, Marsh Lenan’s professional service company launched a system called Sentrisk last year that claims to automatically analyze the shipping manifesto of a company and customs clearance records to create an image of its supply chain. Depending on the customer, Sentrisk relies on large language models to read billions of PDF documents potentially and automatically track separate materials and parts. “This can read something wrong,” says John Davis, a business manager for Sentrisk.
Sentrisk combines this supply chain analysis with data on climate risks in specific locations. “If you want to invest in building a new manufacturing plant, you may be able to choose a place that is less affected by water shortages,” says Davis.
Another challenge is that digital twins need to be constantly updated. “It’s not like the house you build and the house is in this form for 100 years,” he says. “Supply chains change every day.”
And while we have a good logical idea about how climate change on the planet is generally in the coming years, the exact location, timing and size of specific disasters are difficult to predict. This is where new tools for modeling are exposed to climate and severe weather forecast. The semiconductor and AI NVIDIA have a platform called Earth-2 that hopes to address the challenge with the help of other organizations, including the National Oceanic and Atmospheric Office.
The idea is to use artificial intelligence to provide earlier warnings about drought or flood or more accurate prediction of how to grow storms. Some parts of the world only have relatively high information about current climatic patterns. The Earth-2 uses the same type of artificial intelligence that sharpen images in your smartphone camera app to simulate higher resolution data. “This is really useful, especially for small areas,” says Dion Harris, a senior computing manager at NVIDIA.
Companies can feed their data on Earth 2 to improve forecasts even further. They may use this platform to model the effects of climate and climate on specific geography, but the overall range of the project is very wide. “We are building fundamental elements to create a digital twin from Earth,” says Harris.
