9/22/22 | 4:15pm | E51-145

Rene Caldentey

Professor of Operations Management
The University of Chicago

Abstract: We study the interplay between information sharing and order smoothing in the context of a prototypical two-tier inventory management system with one retailer and one supplier. We contribute to this literature by adapting a novel mathematical framework --connecting time series analysis and the theory of Hardy spaces-- that allows us to provide a precise and rigorous answer to the question of when and to what extent there is value in information sharing. Our methodology makes explicit the fact that the problem needs to be cast as both an inventory replenishment problem and an information design problem by exploiting the relationship between the firm's long-term average cost and two distinctive sources of information: (a) the information that is carried by the firm's orders themselves and (b) any additional demand information that the firm is willing and able to share.

A number of important contributions with respect to the existing literature emerge from our approach. For example, we show that for a given inventory replenishment policy there is value in information sharing if and only if the ordering processes is non-invertible with respect to the history of observed demand Furthermore, we derive a fundamental mathematical identity that reveals the value of information sharing by exploiting the canonical inner-outer factorization of the inventory replenishment orders when viewed as an element of the Hardy space H2. We also provide a fully general characterization of invertibility, which is central to our work and that we believe can prove useful in other settings. Finally, our approach fills what we believe is an existing gap in the literature by formalizing and solving the problem of finding an optimal inventory replenishment policy under no information sharing.

This is joint work with Avi Giloni (Yeshiva Univ.), Clifford Hurvich (NYU) and Yichen Zhang (Purdue Univ.)

Bio: René Caldentey is the Eli B. and Harriet B. Williams Professor of Operations Management at The University of Chicago Booth School of Business. His primary research interests include stochastic modeling with applications to revenue and retail management, queueing theory, and finance. He has been published in numerous journals including Advances in Applied Probability, Econometrica, Management Science, Mathematics of Operations Research, M&SOM, Operations Research, and Queueing Systems. He serves on the editorial board of Management Science, Naval Research Logistics and Operations Research.

Prior to joining Booth, Caldentey was a professor in the department of Information, Operations, and Management Science at New York University’s Stern School of Business. Before joining NYU Stern in 2001, he worked for the Chilean Central Bank and taught at the University of Chile.

Professor Caldentey received his Master of Arts in Civil Industrial Engineering from the University of Chile and his Doctor in Philosophy in Operations Management from MIT.