Research at risk: fast delivery of blood transfusions, supplies for troops

ITHACA, N.Y. -- A group of Marines are engaged in combat on a distant island. They are running low on munitions. One of the Marines is perilously wounded and requires a blood transfusion. Whether or not help can reach them in time comes down to one thing: the supply chain.

“Supply chain optimization” may sound cold and technocratic, but for Peter Frazier, his research is ultimately about people.

“It’s about people in trucks and in warehouses, and they’re trying to move stuff around as best they can,” said Frazier, the Eleanor and Howard Morgan Professor of operations research and information engineering in Cornell Engineering. “It’s about Marines who need their equipment in order to survive and be effective. And it’s about the people who make that happen.”

Frazier has been heading up a team, funded by the Office of Naval Research, that builds decision-support tools, optimization methods and artificial intelligence approaches to help the U.S. Navy and Marines transport people and supplies as quickly and effectively as possible in the event of an overseas conflict or humanitarian disaster. The $1.2 million project received a stop-work order in April, just as Frazier’s team was in the middle of supporting a military exercise currently underway in Okinawa in which the Navy is testing the team’s optimization model and providing feedback so the tool can be refined.

The software Frazier’s team is developing can enable optimal transportation to meet a diverse range of needs: food, water, fuel, shelter, munitions, medical supplies and soldiers with special skillsets.

“As academics, we provide this capability at a much lower cost than is charged by many defense contractors. So the government is really getting a good deal by paying us to develop this software,” Frazier said.

That software can enable optimal transportation to meet a diverse range of needs: food, water, fuel, shelter, munitions, medical supplies and soldiers with special skillsets. One of the most vexing items to transport is blood, because it’s highly perishable and demand is never consistent.

“Most of the time you don’t need it, and every once in a while, you really do, and you need it as fast as possible,” Frazier said. “So the decision-making around how much blood to transport is really subtle.”

Frazier’s team includes two doctoral students who are working in collaboration with researchers from University of Wisconsin, University of Southern California, Columbia University, University of California, Berkeley, and Colorado School of Mines, as well as a collection of defense contractors, military information technology specialists and the Third Marine Expeditionary Force. All told, roughly 100 people are involved in the project.

“When the funding stops for the Cornell portion, the other parts of the project try to figure out how to piece things together so things don’t totally fall apart,” Frazier said. “Those teams are also doing other things, so they’re not set up to be able to maintain and improve the code base that we developed.”

To keep the project on track, the Office of Naval Research has used an alternate funding mechanism to keep Frazier and his students working as contractors through the summer. But that is only a temporary fix.

“Here at Cornell, we have a great deal of expertise in the development of the core technology that we use within the project,” Frazier said. “There’s just a small group of people who have the ability to do this.” 

-30-