Assortment Optimization Under Multiple-Discrete Customer Choices

Abstract

We consider an assortment optimization problem where the customer may purchase multiple products and possibly more than one unit of each product. Methodology/results: We adopt the customer consumption model based on the multiple-discrete-choice (MDC) model proposed by Huh and Li (2021). We characterize the performance of revenue-ordered assortments. We show that assortment optimization is NP-hard in general. We present an algorithmic framework that delivers near-optimal algorithms for different variations of the assortment problem, such as unconstrained problem, space and cardinality constrained problem, and the joint assortment and discrete pricing problem. The framework can be fully extended to the mixture of MDC models. Managerial implications: This study provides theoretical foundation and practical guidance for businesses looking to make more realistic decisions about product assortment by incorporating multi-option-multi-unit (MOMU) customer consumption behavior. We also propose several approaches to incorporate uncertainty into the MDC model and estimate the model from customer choice data for assortment purposes.

Keywords: assortment optimization, multiple purchases, approximation algorithm, estimation