Project will strengthen global supply chains with AI

A new grant will fund Cornell research into developing machine learning and blockchain tools to better understand and support the global supply chain, whose vulnerabilities were exposed by the COVID-19 pandemic.

Andreea Minca, associate professor of operations research and information engineering in the College of Engineering, was awarded a three-year, $250,000 grant by Axa, the multinational insurance company based in France, for the research.

“In March we were experiencing shortages, especially in some medical devices, because everything had broken down – there was really a collapse of the supply chains,” said Minca, whose usual area of research is in finance and systemic risk.

“This funding will allow us to develop mathematical technology methodology to identify vulnerabilities in global supply chains and in their insurance/reinsurance networks,” she said, “so those systems become more robust.”

Reinsurance is pooled insurance purchased by insurance companies to protect themselves from large losses stemming from major events. When reinsurers cover primary insurers of the supply chain system, as well as other reinsurers, it creates a complex system; poor design of these systems can lead to catastrophic losses.

Minca’s project aims to develop tools for machine learning-based market designs in supply chain networks with reinsurance contracts, in order to identify potential gaps before another pandemic or disaster occurs.

Because these systems are so complex, machine learning is helpful in identifying the structures that could pose systemic risk, Minca said.

Her team will also use mathematical tools to study the design of system insurance, in which entities such as local governments are incentivized to contribute to a “rainy-day fund” that could buffer supply chains in the event of disasters.

Thirdly, the project will design blockchain systems that use geographic information to verify maritime data, to ensure that this data is consistent and accurate across different providers.

“How do you design a system in which people are reporting the location of a shipment, in a context where there are severe restrictions on where the shipments can go because of a pandemic?” Minca said. “You could have closed ports, you could have outbreaks in some areas, and those become no-go zones – all of that needs to be verifiable by multiple parties.”

The research, Minca hopes, will help supply chains and the companies that insure them better prepare for future catastrophes with worldwide consequences.

“What was really shocking with COVID-19 was this disjointed supply and demand,” she said. “Borders were shutting down, which made it impossible to make and route shipments in an optimal way. To be prepared for that, you need to have some contingency plans, and they need to be financially supported, so things don’t break down at the next shock.”

/Public Release. The material in this public release comes from the originating organization and may be of a point-in-time nature, edited for clarity, style and length. View in full here.