1、Dynamic Pricing under Lost Sales
2、Operational Data Analytics: A Newsvendor Application
1、ProfessorJ. George Shanthikumar，Purdue University
2、Professor Qi Annabelle Feng，Purdue University
1. ProfessorJ. George Shanthikumaris the Richard E. Dauch Distinguished Chair in Manufacturing and Operations Management at Krannert School of Management, Purdue University. He served as the president of POMS and is currently a department editor of Management Science. His research interests are in integrated interdisciplinary decision making, model uncertainty and learning, production systems, queueing theory, reliability, scheduling, semiconductor yield management, simulation and stochastic processes. He has published over 300 papers on these topics and 3 books—Stochastic Models of Manufacturing Systems (with John A. Buzacott) and Stochastic Orders and Their Applications and Stochastic Orders (with Moshe Shaked). He has extensively consulted for various companies such as IBM, Intel and Fujitsu. He is an advisory consultant for Sensor Analytics and a member of the technical advisory board of Inter Molecular Inc. and Reel Solar, Inc.
2. ProfessorQi Annabelle Fengis the John and Donna Krenicki Chair in Operations Management at Krannert School of Management, Purdue University. Her main research interest lies in studying firms’ sourcing decisions in the broad context of supply chain management. Her work focuses on individual firm’s procurement planning in uncertain environment and multiple firms’ interactions in sourcing relationships. She is currently a Department Editor of Data Science, Stochastics and Optimization for Production and Operations Management, the Department Editor of Supply Chain Management for Flexible Manufacturing and Service Journal, and an Associate Editor for Management Science. She received the first prize in the INFORMS Junior Faculty Paper Competition in 2009, the Franz Edelman Award in 2009, the Wickham Skinner Early-Career Research Accomplishment Award in 2012, and the Wickham Skinner Best Paper Award in 2018.
1. Inventory-based pricing under lost sales is an important, yet notoriously challenging problem in the operations management literature. Since the first attempt to solve this problem in the mid-50's, the existing analyses are still limited to either single-period or long-term stationary models with restrictive assumptions on the price-demand relationship. We start with summarizing the latest developments on the single-period problem, pointing out the limitations and proposing a new approach to tackle the problem. In particular, we show that, under very general conditions on the stochastic demand function, the objective function is concave along the optimal price path provided that the price is decreasing in the post-order inventory level. The concavity of the single-period problem, however, does not imply the concavity of the multi-period problem. Using properties of stochastic functions, we derive a set of conditions on the demand function under which the dynamic problem is concave along the optimal price path whenever the optimal price is decreasing in the inventory level. A decreasing price path, though not always optimal, is practically appealing and intuitive to implement. In the case when the optimal price is not monotone in the post-order inventory level, we identify a bounded set of decreasing price paths, along which the objective function is concave. Any decreasing price path outside of this set would lead to a lower expected profit than some path within the set. Our extensive numerical analysis suggests that the restriction of decreasing price path does not lead to a significant optimality gap—The optimal price path is indeed decreasing in most instances and, even when it is not, the profit loss is very marginal.
2. With the development of computing technology and data availability, an increasing attention is paid to data-driven and data-integrated decision making in practice and research. We propose a framework of Operational Data Analytics (ODA) and demonstrate its application through the example of newsvendor model. In this talk, we focus on the situation where we have full structural knowledge but may be uncertain about the statistical characterization of the model. The ODA framework integrates data into decision making by carefully formulating the data-integration model and the validation model. The data-integration model identifies an appropriate class of operational statistics (i.e., the decision as a statistics of the data) and the validation model finds the best performer among the operational statistics. In contrast to the traditional estimation-optimization approach, which does not optimize the actual performance measure, the ODA approach significantly improves decision quality especially when the data is limited.