The Exploration-Exploitation Trade-off in the Newsvendor Problem

69 Pages Posted: 26 Nov 2014 Last revised: 27 Jun 2019

See all articles by Omar Besbes

Omar Besbes

Columbia Business School - Decision Risk and Operations

Juan Chaneton

Columbia Business School - Decision Risk and Operations

Ciamac C. Moallemi

Columbia Business School - Decision Risk and Operations

Date Written: May 25, 2017

Abstract

When an inventory manager attempts to construct probabilistic models of demand based on past data, demand samples are almost never available: only sales data can be used. This demand censoring introduces an exploration-exploitation trade-off as the ordering decisions impact the information collected. Much of the literature has sought to understand how operational decisions should be modified to incorporate this trade-off. We ask an even more basic question: when does the exploration-exploitation trade-off matter? To what extent should one deviate from a myopic policy that takes the optimal decision for the current period without consideration for future periods? We analyze these questions in the context of a well-studied stationary multi-period newsvendor problem in which the decision-maker starts with a prior on a vector of parameters characterizing the demand distribution. We show that, under very general conditions, the myopic policy will almost surely learn, in the long run, the optimal decision one would have taken with knowledge of the unknown parameters. Furthermore, we analyze finite time performance for a broad family of tractable cases. Through a combination of analytical parametric bounds and exhaustive exact analysis, we show that the myopic optimality gap is negligible for most practical instances.

Keywords: demand censoring, inventory management, dynamic learning, finite time analysis, newsvendor, myopic policy, exploration-exploitation trade-off, Bayesian analysis

Suggested Citation

Besbes, Omar and Chaneton, Juan and Moallemi, Ciamac C., The Exploration-Exploitation Trade-off in the Newsvendor Problem (May 25, 2017). Columbia Business School Research Paper No. 14-61. Available at SSRN: https://ssrn.com/abstract=2530653 or http://dx.doi.org/10.2139/ssrn.2530653

Omar Besbes (Contact Author)

Columbia Business School - Decision Risk and Operations ( email )

New York, NY
United States

Juan Chaneton

Columbia Business School - Decision Risk and Operations ( email )

New York, NY
United States

Ciamac C. Moallemi

Columbia Business School - Decision Risk and Operations ( email )

New York, NY
United States

HOME PAGE: http://moallemi.com/ciamac

Register to save articles to
your library

Register

Paper statistics

Downloads
311
Abstract Views
1,728
rank
100,055
PlumX Metrics