Pelin Pekgün
Moore School of Business, University of South Carolina

Estimating Personalized Demand with Unobserved No-purchases using a Mixture Model: An Application in the Hotel Industry

Understanding the "true demand" of customers for a product is of critical importance in many industries to be able to offer the right product to the right customer. For any firm that offers a portfolio of products to its customers, the key problem becomes estimating a choice model that can predict which product a customer is likely to choose to purchase from this portfolio. Over the last decade, the travel and hospitality industries have come to understand that each customer is unique, and the one-size-fits-all policy may not be effective. Accordingly, being able to predict the choices of a heterogeneous customer population and customize product offerings based on individual preferences has become a major differentiator, which requires an accurate estimation of personalized demand as input.

In this study, we develop a novel personalized demand modeling framework using an integrated approach that combines discrete choice modeling under unobservable no-purchases with a data-driven identification of customer segments. In collaboration with our industry partner, Oracle Hospitality Global Business Unit, we demonstrate our methodology in the hotel industry setting, where increased competition has driven hoteliers to look for more innovative revenue management practices such as personalized offers for their guests. Our methodology predicts demand for multiple types of hotel rooms based on guest characteristics, travel attributes, and room features. Our framework combines clustering techniques with choice modeling to develop a mixture of multinomial logit discrete choice models and uses Bayesian inference to estimate model parameters. In addition to predicting the probability of an individual guest's room type choice, our model delivers additional insights on segmentation with its capability to classify each guest into segments (or a mixture of segments) based on their characteristics.

To demonstrate our model's prediction performance, we first run computational experiments using a synthetic dataset with known parameters derived based on a real hotel dataset. Monte-Carlo simulations show that our method gives promising results, outperforming several benchmarks in prediction accuracy, with nearly unbiased estimates of the choice model parameters and the size of no-purchase incidents. We then apply our method on a real hotel dataset and illustrate how the model results can be used to drive insights for personalized offers and pricing. Our proposed framework, thus, provides a practical approach for a complicated demand estimation problem and can help hoteliers profile their guests based on their preferences, which can serve as a valuable input for personalized offer selection and pricing decisions.

Bio:

Pelin Pekgün is an associate professor of management science in the Darla Moore School of Business at the University of South Carolina. She is also the Faculty Director of the Master of Science in Business Analytics program at the Moore School. She was formerly a visiting assistant professor in the Stewart School of Industrial & Systems Engineering at Georgia Tech. She received B.S. and M.S. degrees in industrial engineering from Bogazici University, and M.S. and Ph.D. degrees in industrial and systems engineering from Georgia Tech. Prior to joining academia, Dr. Pekgün led the operations research team at JDA Software's Pricing and Revenue Management Group, where her work with the Carlson Rezidor Hotel Group on stay night price optimization became a finalist for the 2012 INFORMS Franz Edelman Award and the 2013 INFORMS Revenue Management and Pricing Section Practice Award.

Dr. Pekgün's research interests include applications of management science and operations research in pricing and revenue management, supply chain management, marketing/operations interface, and health and humanitarian operations. Her research has been published in journals such as Manufacturing & Service Operations Management (M&SOM), Production and Operations Management, Journal of Operations Management, Decision Sciences, INFORMS Journal on Applied Analytics (IJAA) and IIE Transactions. Dr. Pekgün serves as an associate editor in M&SOM, IJAA and Decision Sciences journals, and is actively involved in INFORMS and POMS societies, currently serving as the VP of Membership and Professional Recognition in the INFORMS Board of Directors.

Please click here to join. 

Meeting ID: 969 4557 2901
Passcode: 335897