Assortment Optimization in the Presence of Focal Effect: Operational Insights and Efficient Algorithms

• 成果类型: Article; Early Access
• 署名作者: Jiang, Bo; Wang, Zizhuo; Xue, Chenyu; Zhang, Nanxi
• 署名单位: Shanghai University of Finance & Economics; Shanghai University of Finance & Economics; Shanghai University of Finance & Economics; The Chinese University of Hong Kong, Shenzhen; East China University of Science & Technology; Western University (University of Western Ontario)
• 刊物名称: PRODUCTION AND OPERATIONS MANAGEMENT
• ISSN/ISSBN: 1059-1478
• DOI: 10.1177/10591478251368435
• 发表日期: 2025
• 关键词: Focal Effect; Assortment Optimization; Focal Luce Model

摘要

The assortment provided by the seller can influence customers' evaluation of item utility. A possible consequence is that certain items in an assortment become focal items to customers, and customers over-evaluate their utilities. We refer to this phenomenon as the focal effect. Kovach and Tserenjigmid (2022) recently propose a focal Luce model (FLM) to describe customers' choices in the presence of the focal effect. The merit of the FLM lies in its flexibility to model different consumer psychology, which leads to varying choice behaviors. In this paper, we use the FLM to capture several scenarios where the focal effect occurs and consider the associated assortment optimization problems. In the first scenario, the focal effect arises from item ranking, and customers prefer items that appear at certain positions in the ranking. This scenario captures the case where customers have a preference for the cheapest items in the assortment, as well as the compromise effect. The second scenario describes the case where the focal effect exists on some predetermined items. An example of this scenario is choice overload, where customers are more likely to choose the no-purchase option when the assortment size is larger. We characterize the optimal assortment structure under each scenario and give the underlying operational insights. We find that the assortment optimization problems under these two scenarios can be solved in polynomial time with some practical assumptions. The polynomial-solvability extends even to the more challenging joint assortment and pricing optimization problems. Finally, we conduct numerical experiments to evaluate the FLM's performance using both synthetic and real datasets. The results show that the FLM performs well in predicting customer choice behavior and the corresponding optimal assortment generates higher profit than benchmark assortments when the focal effect exists.