Amazon Postdoctoral Scientists apply Operations Research on problems in the real world

When customers shop at the Amazon store and learn that a product is sold out, how likely are they to replace it with a similar product from another brand? What are the fastest, most fuel -efficient ways to deliver orders to customers?

If we claim to work on problems in the real world, it is important to actually go out there and work on these problems, in order to realize soil research.

These are among the questions that Chamsi Hssaine and Hanzhang Qin, the initial post -doctoral researchers with Amazon Supply Chain Optimization Technologies (SCOT), explored as they enter the Amazon Postdoctoral Science program in 2022. -Candidates with an opportunity to win industry experience, apply their topic -expertise and teacher from Amazon scient.

“If we claim to work on problems in the real world, it is important to actually go out there and work with thesis problems, for soil research in reality,” said Hssaine, who received a PhD. in Operations at Cornell University and recently joined the data sciences and operations department in southern California’s Marshall School of Business as assistant professor.

Operations Research at Amazon

Howzon’s Team for Supply Chain Optimization Technologies has evolved over time to meet a challenge with staggering complexity.

Qin received a Ph.D. in Computer Science and Technology from the Massachusetts Institute of Technology and will be assistant professor at the National University of Singapore’s Department of Industrial Systems Engineering and Management this fall. He said that the postdoctoral science program at Amazon opened the eyes to the landscape of the real world’s supply chain problems that were not yet resolved.

“You can’t get a real sense of this problem if you only read papers and articles talking about them,” he said. “When I got into this business and could see the data sets describing these problems, I realized that there are still many very important problem in supply chain management and transport.”

Promoting cooperation with postdocs at Amazon

The Amazon Postdoctoral Science program is a natural development of the company’s efforts to engage in the academic community to facilitate an exchange of ideas between academia and Amazon “without causing a brain drain from universities,” explained Salal Humair, a vice president and distinction Scotti Manager.

Participants listen to a presentation at the National Academy of Engineering's Frontiers of Engineering (FOE) 2022 Symposium in Seattle.

Related content

How Amazon forms a set of initiatives to enable Akademia-based talent to harmonize their passions, life stations and career ambitions.

This commitment started with the Amazon Scholars program, which allows retained and high profiled academics to participate in Amazon in flexible capacity, such as part-time scheme. The program is expanded to Amazon visits academics For preset or early annual academics seeking to apply research methods to complex technical challenges while continuing their university work. Postdoctory Science Program Early-Career Academicics.

“Having the best young talent spent a year on Amazon before embarking on their academic career is a great way to build relationships with the next generation of academic leaders,” said Garrett van Ryzin, a difference researcher at the Scot team that was a leader. “These are early days,” he added, “but I trust it is gooir to be very valuable.”

Operating research and optimization

The field of operations and optimization research was unknown to both Hssaine who grew up in Los Angeles, and Qin, which grew up in China until they entered university. But both loved numbers and gravity against math and computer science courses in college. There they discovered both operations that were in line with their individual interests.

Chamsi Hssaine recently joined data science and operations department at the University of Southern California’s Marshall School of Business as assistant professor.

“It was the first time I realized that you could create really elegant mathematical models to solve problems in the real world,” said Hssaine, who learned about the field during an initial engineering race while undergrading at Princeton University. “It really spoke to me.”

She had a major in driving and financial technology in Princeton and attended the Master’s School at Cornell University. Her dissertation focused “on algorithm and incentive design for smart social systems,” she said. “In particular, my research incorporates more realistic models of behavior under incentives and seeks to understand the effects of political decisions.”

For example, a project explored the intersection of how customers decide where to buy certain products and how companies praise these products.

“There are a number of ways in which customers make decisions between company A and company B. My work is trying to understand how different customer behavior assumptions affect this kind of price decision in a complete landscape,” she said.

Hanzhang Qin will be assistant professor at the National University of Singapore’s Department of Industrial Systems Engineering and Management in the fall.

Qin has a major in mathematics and industrial technology at Tsinghua University in Beijing. As part of these studies, he was exhibited for operational research and stays in focus on it while he was at MIT, where he received a master’s degree in electrical engineering and computer science and another master’s degree program in transport.

Then he pursued a Ph.D. in computer science and engineering with a primary focus on areas of operations that use and probability to navigate uncertainty.

For example, an area of ​​his research focused on developing a common pricing and inventory management system for times when demand is uncertain. Another interest was to develop routes for delivery vehicles before demand is known.

“When you plan routes in advance, some of the roads are overlapped on some drivers intentionally so they can help each other and coordinate on these overlap routes,” Qin said. He studied the value of this overlap in routes and found that “a very little love for overlap can mean the system’s performance.”

Postdoctorical science

Asy wrapped their PhD research, both HSSAINE and QIN secured employment positions in Academy. Still, both chose to postpone their appointments for a year to gain industry experience.

“In particular, Amazon seemed like a natural fit for my research because the opportunity to use my methodological toolbox to Scots rich problem space,” Hssaine said. “And Amazon had been on my radar because I did an internship at Amazon in my third year of PhD” ”

Hssaine’s Hand Project is on in -depth optimization – coordination of where suppliers and sellers send their product into the Amazon network. This involved building models that explore, among other things, the exchange between measurements as the closest stock to the seller or the seller and the levels of overload on these warehouses.

Animation shows a map of the United States and each of the 8 individual regions that are the result of Amazon's regionalization

Related content

The pandemic turbo-charged retail growth teams for researchers at Amazon threw a path forward to deal with the scale.

For example, if the warehouse closest to the seller is overload, it is the overload caused by the product to a customer. Sending a shipment to an overloaded stock will also have knock-on effects for other products and customers.

“When you think about where to send a shipment, you don’t just think about the costs that incur the costs it imposes on the rest of the system,” Hssaine said.

This research required to find data that is often hidden from ordinary sight, now van Ryzin. For example, there is not a long tail of trucks in stocks signaling overload. On the contrary, sellers are planning delivery agreements and overloading means that the next available agreement may be the following week. It emerges as a delay.

“She had to do a lot to dig around to find that tail was really where it was manifested, and we even have visibility about how to get huts,” Van Ryzin said.

Qin’s research at Amazon under Humair took two tracks. One explored ways to improve the algorithms used to sell excess storage through multiple channels, such as Markdowns in the Amazon store and targeted advertising on another site.

This image is an overhead image inside an Amazon center, the workers can be seen moving in the middle of hundreds of boxes sitting on conveyor belts and wagons, in the upper left foreground, a yellow railing developments into.

Related content

Howzon’s researchers developed a first-stroke multi-echelon system for inventory purchases and location.

“This is a relatively immature area in operational research,” Human noted. “There are several ways we can make the products more attractive.”

In another project, Qin used his Ph.D. -Research in planning effective ways for prose, storage and road stock for customers. The preliminary research focused specifically on modeling compromise between carbon emissions, stock levels at fulfillment centers and delivery routes and can eventually inform the company’s progress towards its climate relationship.

Qin presented his delivery route planning study at the SIAM conference on optimization in Seattle this spring. Hssaine presented his work with incoming optimization at the same conference.

Back to Akademia

As Hssaine and Qin enter the next phase of their careers in the academy, they will build on the research conducted in the past year on Amazon, taking into account what they learned about the types of questions that decision makers need.

“As academics we can be so divorced from it,” said Hssaine. “Although what I have been working on at Amazon is related to the kinds of things I was thinking about Cornell, it has allowed me to see a much wider rank.”

Ozge Sahin

Academics at Amazon

Johns Hopkins Business School professor and Amazon Scholar focuses on improving customer experiences.

Qin, who has worked with several companies throughout his academic career, will take with him a newly won appreciation of Amazon’s “Bias for Action” management principle that appreciates speed in business.

“It’s much more effective,” he said of researching Amazon. “This experience has helped me to become comfortable with the quick work.”

Humair and Van Ryzin expect the exchange of ideas with their first class by Amazon Post -doctorate scientists will continue Asy to start their careers in academia. For example, both Qin and Hssaine are working on research articles with Amazon colleagues.

More broadly, Humair believes that the community experience will help Qin and Hssaine focus their academic research on topics that have effect in the real world.

“As academics, you have a great flexibility on what you choose to work on,” he said. “What I hope they take away is the judgment of what is very important problem to work on.”

Leave a Comment