Details

Yunan Liu, Ph.D.
Associate Professor
Department of Industrial and Systems Engineering, North Carolina State University

Abstract: Optimal Pricing and Capacity Sizing in Service Systems with Online Demand Learning

In this talk, we demonstrate how to apply robust reinforcement learning techniques (an area of machine learning, alongside supervised and unsupervised learning) to classical operations research problems. Specifically, we consider a multi period pricing and staffing problem in a service queueing system with an unknown demand curve. The service provider’s objective is to dynamically adjust the service price p (and service rate µ) so as to maximize cumulative expected revenues (the sales revenue minus the delay penalty) over a given finite time horizon; in doing so, the service provider needs to resolve the tension between learning the unknown demand curve λ(p) and maximizing earned revenues. Using a stochastic-gradient-based method, we develop an online reinforcement learning algorithm that balances between exploitation (of unknown demand curve) and exploration (of short-term feedback of candidate solutions). We provide asymptotic bounds for the regret which benchmarks the case when λ(p) is unknown.