A Dynamic Clustering Approach to Data-Driven Assortment Personalization

Abstract

AboutSectionsView PDF ToolsAdd to favoritesDownload CitationsTrack CitationsPermissionsReprints ShareShare onFacebookTwitterLinked InEmail Go to Section HomeManagement ScienceVol. 65, No. 5 A Dynamic Clustering Approach to Data-Driven Assortment PersonalizationFernando Bernstein , Sajad Modaresi, Denis SauréFernando Bernstein , Sajad Modaresi, Denis SauréPublished Online:5 Sep 2018https://doi.org/10.1287/mnsc.2018.3031AbstractWe consider an online retailer facing heterogeneous customers with initially unknown product preferences. Customers are characterized by a diverse set of demographic and transactional attributes. The retailer can personalize the customers’ assortment offerings based on available profile information to maximize cumulative revenue. To that end, the retailer must estimate customer preferences by observing transaction data. This, however, may require a considerable amount of data and time given the broad range of customer profiles and large number of products available. At the same time, the retailer can aggregate (pool) purchasing information among customers with similar product preferences to expedite the learning process. We propose a dynamic clustering policy that estimates customer preferences by adaptively adjusting customer segments (clusters of customers with similar preferences) as more transaction information becomes available. We test the proposed approach with a case study based on a data set from a large Chilean retailer. The case study suggests that the benefits of the dynamic clustering policy under the MNL model can be substantial and result (on average) in more than 37% additional transactions compared to a data-intensive policy that treats customers independently and in more than 27% additional transactions compared to a linear-utility policy that assumes that product mean utilities are linear functions of available customer attributes. We support the insights derived from the numerical experiments by analytically characterizing settings in which pooling transaction information is beneficial for the retailer, in a simplified version of the problem. We also show that there are diminishing marginal returns to pooling information from an increasing number of customers.The online appendix is available at https://doi.org/10.1287/mnsc.2018.3031.This paper was accepted by Gad Allon, operations management. Previous Back to Top Next FiguresReferencesRelatedInformationCited byAnalytics Applications in Fashion Supply Chain Management—A Review of Literature and PracticeIEEE Transactions on Engineering Management, Vol. 70, No. 4Market Segmentation TreesAli Aouad, Adam N. Elmachtoub, Kris J. Ferreira, Ryan McNellis10 March 2023 | Manufacturing & Service Operations Management, Vol. 0, No. 0Data-Driven Revenue Management: The Interplay of Data, Model, and DecisionsNingyuan Chen, Ming Hu10 March 2023 | Service Science, Vol. 0, No. 0Data Aggregation and Demand PredictionMaxime C. Cohen, Renyu Zhang, Kevin Jiao7 July 2022 | Operations Research, Vol. 70, No. 5Estimating Personalized Demand with Unobserved No-Purchases Using a Mixture Model: An Application in the Hotel IndustrySanghoon Cho, Mark Ferguson, Pelin Pekgün, Andrew Vakhutinsky30 March 2022 | Manufacturing & Service Operations Management, Vol. 0, No. 0Cross-Category Retailing Management: Substitution and ComplementarityChenxu Ke, Ruxian Wang2 December 2021 | Manufacturing & Service Operations Management, Vol. 24, No. 2Applied Machine Learning in Operations Management11 May 2021Discrete Choice Models and Applications in Operations ManagementRuxian Wang18 October 2021Assortment Rotation and the Value of ConcealmentKris Johnson Ferreira, Joel Goh27 July 2020 | Management Science, Vol. 67, No. 3Retail Analytics: The Quest for Actionable Insights from Big Data on Consumer Behavior and Operational ExecutionSSRN Electronic Journal, Vol. 65What is the right production strategy for horizontally differentiated product: Standardization or mass customization?International Journal of Production Economics, Vol. 223Data Analytics in Operations Management: A ReviewVelibor V. Mišić, Georgia Perakis19 September 2019 | Manufacturing & Service Operations Management, Vol. 22, No. 1Product Assortment Policy in Business Entities: Tactics and Strategies1 January 2020 | International Journal of Innovation and Economic Development, Vol. 6, No. 2Price Discounts and Customized Product Assortment under Multinomial Logit Choice Model: A Robust ApproachSSRN Electronic Journal, Vol. 17Personalized Sales Targets with Customer ChoicesSSRN Electronic Journal, Vol. 5Estimating Personalized Demand with Unobserved No-purchases using a Mixture Model: An Application in the Hotel IndustrySSRN Electronic Journal, Vol. 57Privacy-Preserving Personalized Revenue ManagementSSRN Electronic Journal, Vol. 105Applied Machine Learning in Operations ManagementSSRN Electronic Journal, Vol. 67Data Analytics in Operations Management: A ReviewSSRN Electronic Journal, Vol. 17Data Aggregation and Demand PredictionSSRN Electronic Journal, Vol. 65Maximizing the Potential of Targeted Marketing: A General Framework for Customized Category Promotions in RetailSSRN Electronic JournalThreshold Utility Model with Applications to Retailing and Discrete Choice ModelsSSRN Electronic Journal, Vol. 65The Use of Binary Choice Forests to Model and Estimate Discrete Choice ModelsSSRN Electronic Journal, Vol. 65Online Assortment Optimization with High-Dimensional DataSSRN Electronic JournalLearning from Clickstream Data in Online RetailSSRN Electronic Journal, Vol. 65Assortment Rotation and the Value of ConcealmentSSRN Electronic Journal, Vol. 67Learning in Combinatorial Optimization: What and How to ExploreSSRN Electronic Journal, Vol. 33 Volume 65, Issue 5May 2019Pages 1949-2443 Article Information Supplemental Materials Metrics Information Received:April 18, 2016Accepted:December 13, 2017Published Online:September 05, 2018 Copyright © 2018, INFORMSCite asFernando Bernstein, Sajad Modaresi, Denis Sauré (2018) A Dynamic Clustering Approach to Data-Driven Assortment Personalization. Management Science 65(5):2095-2115. https://doi.org/10.1287/mnsc.2018.3031 Keywordsdata-driven assortment planningpersonalizationdynamic clusteringmultiarmed banditPDF download