【Time】10:00-11:00, July 17, 2015 (Friday)  

【Venue】 North 510, Shunde Bldg. 

【Title】A Data-Driven Approach for the Joint Pricing-Inventory Problem

【Speaker】Dr. Mengshi Lv , Assistant Professor of Operations Management at the Krannert School of Management, Purdue University

【Host】Dr. Simin Huang

【Abstract】We present a data-driven approach for the joint pricing-inventory problem where the relationship between price and demand is unknown. The approach does not make parametric assumptions on the underlying demand model. Instead, it is based on historical observations and basic domain knowledge, which are available in practice. Parametric programming is applied to efficiently estimate the conditional quantile path of the demand. Smoothing and kernelization are used to improve the estimates and decisions. Additional domain knowledge, such as demand concavity, can also be incorporated in the approach. Numerical experiments show that the data-driven approach is able to find close-to-optimal solutions. Smoothing, kernelization, and the incorporation of additional domain knowledge can significantly enhance the performance of the approach.

【Bio】Mengshi Lv is an Assistant Professor of Operations Management at the Krannert School of Management, Purdue University. He received his PhD in Industrial Engineering and Operations Research in 2014 at the University of California, Berkeley. His current research interests include data-driven and robust supply chain management, cloud computing, and business analytics. He has worked for Oracle America Inc., HP Labs, and Google Inc., in the area of demand analytics and supply chain management. He graduated from Tsinghua University in 2009 with a BEng and an MEng in Industrial Engineering.