Paper Co-Authored by Pelin Pekgun Forthcoming in Production and Operations Management

A new paper co-authored by Thomas H. Davis Professor in Business and Professor of Analytics Pelin Pekgun is forthcoming in Production and Operations Management, the flagship journal of the Production and Operations Management Society, which represents the interests of production and operations management (POM) professionals worldwide.

Written alongside Sanghoon Cho from Texas Christian University, Jongho Im from Yonsei University and Mark Ferguson from the University of South Carolina, “Robust Demand Estimation with Customer Choice-Based Models for Sales Transaction Data” explores a novel method for predicting customer choice behavior in the context of censored demand estimation.

“In this paper, we focus on the censored demand estimation problem, prevalent in many industries such as hotels, airlines, and retail, where the firm cannot directly observe customers who choose not to purchase any product,” said Pekgun. “Our method combines several desirable properties, which makes it a better fit for environments, where the available choice sets or attributes of the products in the choice sets change over time.”

Through the use of Monte-Carlo simulations, the team was able to show that this new approach provides promising predictions of customer choice behavior when compared with other generally used methods, while also clearly outperforming those methods in scenarios where the product prices change frequently over time.

The team was further able to illustrate the improved estimation accuracy of their method compared to benchmark methods by utilizing a real hotel transaction dataset provided by Oracle Labs.

“We are grateful to our industry partner, Oracle Labs, for supporting our research with a rich data set and valuable insights,” said Pekgun.

“Our new estimation method offers valuable tools for researchers and practitioners to more accurately estimate demand using discrete choice models, especially in settings employing dynamic pricing and personalized offering practices. As such, this method is particularly well suited for applications in online retail, travel, transportation, and hospitality.”

Covering a wide range of topics in product and process design, operations, and supply chain management, Production and Operations Management is recognized as one of the top journals in its field, appearing in both the Financial Times and University of Texas at Dallas’ list of top journals.