Xiuli Chao, Ph.D.University of Michigan, Industrial and Operations Engineering
Dynamic optimization in supply chain management is usually based on information about distribution of random customer demand, and customer responses to selling prices, etc. In this talk, we will discuss the scenario where such information is not available a priori, and has to be learned on the fly. We develop data-driven learning algorithms for several classes of inventory control and pricing optimization problems that actively integrate exploration and exploitation, and converge to the true (but unknown) optimal solution at the fastest possible rate. Numerical results show that the algorithms are quite effective. This talk is based on several joint works with students and colleagues.
Xiuli Chao is a professor in the Department of Industrial and Operations Engineering at the University of Michigan. His research interests include queueing, scheduling, inventory control, and supply chain management, and their many applications in manufacturing and service systems. Xiuli is the co-author of two books, “Operations Scheduling with Applications in Manufacturing and Services” (Irwin/McGraw-Hill, 1998), and “Queueing Networks: Customers, Signals, and Product Form Solutions” (John Wiley & Sons, 1999), and he is the co-developer of Lekin scheduling system. Xiuli received the 1998 Erlang Prize from the Applied Probability Society of INFORMS, and he also received the 2005 David F. Baker Distinguished Research Award from the Institute of Industrial and Systems Engineers (IISE, formerly IIE). Xiuli received his doctoral degree in Operations Research from Columbia University.
Host: Dr. Lin Li