Working papers
M. Tekin, “Not All Stars Align: Cross-platform Discrepancies in Online Restaurant Reviews and the Risk of Misaligned Discounts”
G. Yorgancilar, M. Tekin, O. Hazir, T. Kucukyilmaz "The Grouping Genetic Algorithm for Workload Smoothing in U-Shaped Assembly Lines"
M. Tekin, K. Talluri, “Popular location estimation for restaurants using public data”
K. Talluri, M. Tekin, “Optimal location for competing retail service facilities”
Summary: This research offers a strategic solution for the challenges associated with retail location decision-making, focusing on restaurant industry. The difficulty lies in predicting demand and popularity without access to critical data, both presently and in the future, as historical demand data is absent, and competitors' information is sensitive. Our approach integrates publicly available demographic and density data with online review metrics from Yelp. We develop an estimation methodology that allows us to discern preferences of the population for various cuisines, price ranges, and other restaurant attributes across the city. Subsequently, we build an optimization model to assist firms in evaluating the viability and profitability of different locations. This framework provides a practical toolkit, enhancing a firm's chances of survival in competitive markets. Notably, our findings reveal nuances in customer behavior, including sensitivity to distance based on restaurant price ranges and diverse impacts of review ratings across restaurant types.
Early stage presentation of the project at Cognitive Analytics Management conference organized by American University of Beirut is available on YouTube, 2016.
Refereed Publications
A. Li, K. Talluri, M. Tekin, "Estimating demand with unobserved no-purchases on revenue-managed data" M&SOM, 2025
K. Talluri, M. Tekin, “Estimation using marginal competitor sales information” Journal of Operations Management, 2025.
Summary: We tackle the estimation problem under a market-share model, focusing on the hotel industry, and develop methodologies to overcome the following significant challenges: (i) competitor’s data is aggregated across multiple LOS with distinct demands (ii) we do not observe no-purchasers, i.e. those who purchase neither ours nor the competitor’s products, and finally, (iii) the competitor makes private sales to groups before the retail sales period; thus even the competitor’s capacity is unobservable.
M. Tekin, S. Ozekici, "Mean-Variance Newsvendor Model with Random Supply and Financial Hedging", IISE Transactions 47, 1-19, 2015.
PhD thesis: Competitive Intelligence Analytics [pdf].
MS thesis: Risk-Sensitive Approach to Inventory Management [archived online].