4/8/2016 Emily Scott
“Although my research primarily focuses on supply chain, I try to learn other techniques from different communities, such as machine learning...”
Written by Emily Scott
When Linwei Xin first came to the United States, he was pursuing his doctorate degree in pure mathematics from the Georgia Institute of Technology. But after two years, he decided to switch his research area to operations research. He said this is one of the best decisions he has made in his life.
“I had studied math but I wanted to see how to apply this knowledge to solve some real-world problems,” he said.
Without this decision to study operations research, he wouldn’t have eventually been able to join ISE faculty as an assistant professor in the fall of 2015 after completing his Ph.D. from the Georgia Institute of Technology.
During his Ph.D., Xin achieved what he described as the biggest accomplishment of his career, winning the 2015 George E. Nicholson Student Paper Competition. This competition is held each year to identify and honor outstanding papers in the field of operations research and the management sciences written by a student.
Xin first earned industry experience by working for the IBM T.J. Watson Research Center in 2013 and WalmartLabs in the summer of 2015. At WalmartLabs, he solved an important supply chain problem.
It had to do with a problem that Walmart.com faced in selling items online and deciding when to restock items, and how many items to order. This problem becomes more difficult when importing items from foreign countries and dealing with the lead time — the time in between ordering the product and receiving it. Sometimes, it can take up to 13 weeks to receive a shipment.
In his position as a data scientist, Xin designed solutions and provided theoretical support for a policy to solve this problem. Right now, the policy is in the process of being implemented at Walmart.com.
But it has always been Xin’s goal to become a professor, which is why he was interested in joining faculty in the Department of ISE.
“Teaching in college, I can reach many students,” Xin said. “If you see their improvement, it’s an accomplishment you’ve made.”
He was especially excited to join ISE faculty for the background of supply chain research that the department boasts and for the strength of the College of Engineering as a whole.
“There is a lot of action between ISE and other departments,” he said. “It’s a good opportunity to explore my research area.”
Xin said he enjoys the quietness of the Champaign-Urbana community, which he said is perfect for doing research.
“Also, the Chinese restaurants here are really good,” he said with a laugh.
Currently, Xin’s research is still focused on some ongoing projects that he started for Walmart.com.
One of these projects deals with a problem that the company faced during the holiday season.
The demand for some specific items increased significantly during the season, but after, their demand decreased sharply. Xin, as well as Professor Xin Chen and PhD student Juan Xu is working on making algorithms that will detect at what time this demand changes, which he said is very difficult to do.
Next, he hopes to incorporate machine learning — a type of artificial intelligence that can predict patterns in data — into his research.
“Although my research primarily focuses on supply chain, I try to learn other techniques from different communities, such as machine learning, and then apply those techniques to solve a problem in supply chain,” he said.
In addition to his research, Xin said he enjoys teaching and interacting with students. He is currently teaching a course in stochastic processes and applications, in which he introduces certain stochastic models and teaches how they can help solve problems.
He said one of the main goals of his career is to see his students thrive, graduate and have successful careers.
Another one of his goals is to solve major open problems in the field. He said he hopes to be able to continue to provide theoretical support for algorithms used by companies.
“I’d like to prove theorems, but I’d also like to see how we use those theorems to benefit the world.”