2016 INFORMS Annual Meeting Program

MA48

INFORMS Nashville – 2016

2 - Reining In Onion Prices By Introducing A Vertically Differentiated Substitute: Models, Analysis, And Insights Muge Yayla-Kullu, Associate Professor, University of Central Florida, Orlando, FL, United States, mugeyayla@hotmail.com Omkar D Palsule-Desai, Nagesh Gavirneni We examine the pricing ordeal in India’s onion markets caused by the fresh produce traders. As a remedy, policy makers have been proposing to establish processed produce competition in the market by either cooperatives or private firms. We formulate and analyze this situation in a mathematical model that captures (i) competition between non-profit and for-profit organizations, (ii) consumers’ valuation discount for the processed produce, and (iii) perishability of the fresh produce. We identify and discuss the conditions under which (i) it is optimal to introduce the processed produce; and (ii) the processed onion should be managed by cooperatives instead of private firms. 3 - Impact Of Encroachment When Competing Manufacturers Sell Through A Common Retailer Parshuram Hotkar, Doctoral Student, University of Texas, Austin, University Station, B6500, Austin, TX, 78712, United States, parshuram@utexas.edu, Steve Gilbert We consider a setting in which two manufacturers sell partially substitutable products through a common retailer, and examine the impact of the development of a direct sales channel for one of the manufacturers. We find that the retailer’s ability to purchase from another manufacturer can alter many of the results that have been obtained for how encroachment affects the interactions between a manufacturer and a retailer. In addition, we find that the non-encroaching manufacturer can benefit from his rival’s direct channel. 4 - Effectiveness Of Targeted Return Management On Retailer’s Profitability Tolga Aydinliyim, Assistant Professor, Baruch College, CUNY, New York, NY, United States, Tolga.Aydinliyim@baruch.cuny.edu Mehmet Sekip Altug As retailers offer more flexible return policies, customer abuse and fraudulent returns are also on the rise. In order to combat that situation, instead of changing the return policies for everyone, retailers started to implement a tool that identifies those “renters”. In a price-setting newsvendor framework, we first study the retailer’s uniform return policy in which the retailer offers the same return policy to everyone; then we study a targeted return policy where the retailer identifies the renters segment and offers a different return policy to that segment. We argue how and when targeted return management leads to improvement in retailer’s profitability. Chair: Candace Arai Yano, University of California-Berkeley, IEOR Dept. and Haas School of Business, Berkeley, CA, 94720, United States, yano@ieor.berkeley.edu 1 - Dynamic Pricing And Replenishment Of Vertically Differentiated Products With Customer Upgrades Oben Ceryan, Drexel University, Philadelphia, PA, 19104, United States, oceryan@drexel.edu, Izak Duenyas, Ozge Sahin We study the impact of product upgrades on a firm’s pricing and replenishment policies by considering a multiple period, two-stage model where the firm first sets prices and replenishment levels, and after observing the demand, it decides whether to upgrade any customers to a higher quality product. We characterize the structure of the optimal upgrade, pricing, and replenishment policies and find that offering upgrades assists in preserving the vertical price differentiation between products. 2 - Dynamic Pricing Of Vertically Differentiated Products With Sales Milestones Chi-Guhn Lee, University of Toronto, chi@mie.utoronto.ca, Sajjad Najafi We study the dynamic pricing of multiple substitutable products over a finite horizon subject to sales milestone constraints. Customers are utility maximizer and consider the relative importance between the price and the quality of product. We formulate the problem as a Markov decision process with probabilistic constraints and relax the constraints following the Lagrangian relaxation to apply KKT conditions. The proposed model is specifically suitable for applications in which the achievement of sale targets plays a crucial role for managers such as residential real estate sales and penetration strategy. MA47 209C-MCC Optimizing Pricing for Multiple Substitutable Products Sponsored: Revenue Management & Pricing Sponsored Session

3 - Optimizing Pricing For Multiple Substitutable Products Kevin Li, University of California - Berkeley, kbl4ew@berkeley.edu We address a retailer’s problem of setting prices, including promotion prices, over a multi-period horizon for substitutable products within a category, considering the effects of reference prices on customers’ strategic buying behavior, including stockpiling. We utilize an embedded model in which customers make purchasing and consumption decisions over multiple periods to maximize utility. We present structural results and examples that provide insights into the properties of optimal policies. 4 - Pricing Two Substitutable Products With Limited Demand Information Zhi-Long Chen, Professor, University of Maryland, 691 Market Street East, College Park, MD, 20742, United States, zchen@rhsmith.umd.edu, Ming Chen We consider a practical dynamic pricing problem with two substitutable products involving a number of business rules and constraints commonly seen in practice. There is limited demand information. A case with inter-product substitution only, and a case with both inter-product substitution and intertemporal substitution are studied. We propose DP algorithms for both cases, and for the latter, more general, case, we develop a fully polynomial time approximation scheme. We derive a number of managerial insights. 210-MCC Social Media Analytics for Business Applications Invited: Social Media Analytics Invited Session Chair: Yuheng Hu, University of Illinois-Chicago, 601 S Morgan St, Chicago, IL, 60607, United States, yuhenghu@gmail.com 1 - Content Complexity, Similarity, And Consistency In Social Media: A Deep Learning Approach We investigate the effect of social media content on customer engagement using company-generated posts from Tumblr. We employ state-of-the-art machine learning approaches to extract features from textual and visual sources that effectively capture their semantics. With such semantic representations, we develop novel complexity, similarity, and consistency measures of social media content. The results show that proper visual stimuli, complementary textual content, and consistent themes have positive effects on the engagement, and that content demanding significant concentration levels have the opposite effects. This work shows how unstructured data can be translated into insights. 2 - Does Twitter Sentiment Move Stock Prices: Evidence From An Event Study Of The Amsterdam Exchange Yixin Lu, George Washington University, yixinlu@gwu.edu The tremendous amount of information that accumulates and propagates via social media has profound impact on individual businesses as well as the entire market environment. This research focuses on the impact of twitter sentiment on leading stocks traded on the Amsterdam Exchange. By combining event study and sentiment analysis, we demonstrate that twitter peaks are strongly associated with abnormal returns. However, such association is asymmetric with respect to the valence of the sentiment. 3 - Patient Base And Price Premium For Online Health Consultations Liwei Chen, University of Cincinnati, vivienclw@gmail.com, Arun Rai, Xitong Guo Online health consultation communities (OHCCs) enable physicians to signal professional competence and compassionate care for patients, and allow patients to spread online reviews with peer patients. We examine the interactions between the signaling and online feedback mechanisms that explain how physicians build trust with patients and gain social and economic advantages in healthcare services. We scraped multilevel data and traced physicians biweekly over one year from an OHCC in China. Using mixed effects modeling, we find interesting interaction effects between trustworthiness signals and properties of feedback on online patient base and price premium. MA48 Gene Moo Lee, University of Texas at Arlington, Arlington, TX, United States, gene.lee@uta.edu Donghyuk Shin, Shu He, Andrew B Whinston

137