| Jiankun Sun
美国西北大学凯洛格商学院博士生
|
|
【时间】2019年2月27日（周三）早 10:00
Jiankun Sun is a fifth-year PhD candidate in Operations Management at Kellogg School of Management, Northwestern University. Prior to studying at Kellogg, she received her B.E. in industrial engineering at Tsinghua University. Her research interest is integrating both theoretical modeling and data analytics to study practice-driven problems in supply chain management, inventory management and platform operations. She has been doing her research internship in data analytics at Alibaba Group since summer 2018.
【Speaker】Jiankun Sun, Phd candidate，Kellogg School of Management, Northwestern University.
【Topic】Theory and Practice in Supply Chain Management: From Sourcing to Fulfillment
【Time】Wednesday, Feb 27, 2019, 10:00am
【Venue】Room 453, Weilun Building, Tsinghua SEM
【Organizer】Department of Management Science and Engineering
【Title】 Theory and Practice in Supply Chain Management: From Sourcing to Fulfillment
【Abstract】I will discuss two research journeys that I have in supply chain management: theoretical work in sourcing and data-driven study in fulfillment.
The talk will be mostly focused on my theoretical work in dual sourcing. We provide closed-form solutions to a robust optimization model for inventory management with two supply sources or modes with general lead times. The fast source is more expensive than the slow source. While the optimal stochastic policy for non-consecutive lead times has been unknown for over 50 years, we prove that the optimal robust policy is a dual index, dual base-stock policy that constrains or “caps” the slow order. Optimality is established in a rolling horizon model that can accommodate non-stationary demand. As the lead time difference grows, the capped dual index policy increasingly smooths slow orders and, for stationary demand, converges to the tailored base-surge policy, which places a constant slow order and has been shown to be asymptotically optimal. In an extensive simulation study, the capped dual index policy performs as well as, and can even outperform, the best heuristics presented in the stochastic inventory literature.
However, theories might not work as well as expected when they are implemented in the practice of fulfillment. Therefore, at the end of my talk, I will briefly discuss another project on human’s non-conformance with algorithmic prescriptions in logistics. In many operational processes of warehouse management and logistics, the workers may not follow the algorithmic solutions in execution for various reasons. In this study, we use machine learning techniques to predict human’s discretion behavior and adjust the algorithmic prescriptions accordingly. We run a large-scale field experiment at Alibaba Group on different bin-packing algorithm designs. We find that our new design with human discretion prediction and algorithmic prescription adjustment could improve the conformance and productivity of workers, and potentially has a great economic and environmental impact.
|
