2016 INFORMS Annual Meeting Program

SC49

INFORMS Nashville – 2016

SC48

3 - Dynamic Pricing With Stochastic Reference Price Effects Xin Chen, UIUC, xinchen@illinois.edu, Zhenyu Hu, Yuhan Zhang We study a dynamic pricing problem of a firm facing stochastic reference price effect. Randomness is incorporated in the formation of reference prices to capture exogenous factors that affect consumers’ memory processes. We derive an explicit expression for the optimal pricing strategy which allows us to obtain the distribution of the steady state reference price. We compare the expected steady state reference price to the steady state reference price in a deterministic model and we find that the former one is always higher. A transformation technique is presented to show how one can extend the analysis to higher dimensional problems in which consumers have heterogeneous reference prices. 4 - Hidden City Travel And Its Impact On Airfare: The Case With Competing Airlines Jaelynn Oh, University of Utah, jaelynn.oh@business.utah.edu Tim Huh We study the impact of hidden-city ticketing on airfare pricing in a setting where two airlines compete on a hub-and-spoke flight network. New Topics in Revenue Management Sponsored: Revenue Management & Pricing Sponsored Session Chair: Stefanus Jasin, University of Michigan, Ann Arbor, MI, United States, sjasin@umich.edu 1 - Managing Dynamic Mobile Push Advertisements At Alibaba Van Anh Truong, Columbia IEOR, vt2196@columbia.edu In recent years, e-commerce companies are seeing an increasing amount of transactions completed via mobile platforms, such as apps in iOS and Android systems. In China, the e-commerce market share of a mobile app developed by Alibaba Group, which has been installed on several hundred million devices, is rapidly replacing that of traditional e-commerce markets hosted on webpages. We study the problem of managing the allocation of push notifications sent to users by this app which recommends products tailored to every user. 2 - Efficient Algorithms For Dynamic Pricing Problem With Reference Price Effect Zhenyu Hu, National University of Singapore (NUS), bizhuz@nus.edu.sg We analyze a dynamic pricing model in which demand at each period depends on not only the current price but also past prices through reference prices. A unique feature but also a significant challenge in this model is the asymmetry in reference price effect which implies the underlying optimization problem is non- smooth and no standard optimization methods can be applied. We identify a few key structural properties of the problem, which enable us to develop strongly polynomial time algorithms to compute the optimal prices for several plausible scenarios. We further conduct numerical experiments to study the optimal price path and demonstrate the value of dynamic pricing when demands are seasonal. 3 - Optimal Stopping And Worker Selection In Crowdsourcing: An Adaptive Sequential Probability Ratio Test Framework Xi Chen, Stern, NYU, xchen3@stern.nyu.edu In this talk, we propose an adaptive sequential probability ratio test (Ada-SPRT) that obtains the optimal experiment selection rule, stopping time,and final decision rule under a single Bayesian decision framework. Our motivating application comes from binary labeling tasks in crowdsourcing, where the requestor needs to simultaneously decide which worker to choose to provide the label and when to stop collecting labels to save for budget. We characterize the structure of the optimal adaptive sequential design that minimizes the Bayes risk through log-likelihood ratio statistic and develop dynamic programming based algorithms for both non-truncated and truncated tests. 4 - Dynamic Joint Pricing And Order Fulfillment For E-commerce Retailers Yanzhe Lei, University of Michigan, Ann Arbor, MI, United States, leiyz@umich.edu, Stefanus Jasin, Amitabh Sinha We consider a dynamic joint pricing and order fulfillment optimization problem in the e-commerce retailing context, where a retailer sells a catalog of products to customers from different locations and fulfills orders through multiple fulfillment centers. The objective is to maximize the total expected profits, defined as revenues minus shipping costs. We propose two heuristics that are easily implementable, and show both theoretically and numerically their good performances compared to reasonable benchmarks. SC47 209C-MCC

210-MCC Social Media in Marketing and Talent Management General Session Chair: Fujie Jin, Indiana University, 1309 East Tenth Street, Suite 4100, Bloomington, IN, 47405, United States, jinfujie@wharton.upenn.edu 1 - Social Media Marketing In Product Harm Crisis Shu He, University of Connecticut, shuhe@utexas.edu We conceptualize the dual roles of a firm’s social media strategy—-offensive and defensive marketing strategies—-and study how non-focal firms adjust these two components of strategy in response to a product-harm crisis. We use daily social media activity of 56 major airlines on Twitter around the time of an airplane crash to study how non-focal airlines harness these two functions before and after the crisis. We find that a non-focal airline increases its defensive marketing effort but decrease its offensive marketing effort after the crash. The adjustment of offensive marketing decrease is significantly attenuated by whether the non-focal airline directly competes with the focal airline. 2 - Influence Of Social Media On Flash Sales: An Empirical Analysis Karthik Kannan, Georgia Tech, karthikbabu.nk@scheller.gatech.edu, Jeffrey Hu, Sridhar Narasimhan The emergence of e-commerce platforms has democratized both the production and consumption of goods and services. While anyone can offer their services through these platforms, sellers with little brand recognition have to overcome high search cost faced by prospective buyers in order to succeed in these markets. We study two mechanisms - flash sales and social media - used by sellers to promote their products by collecting sales and social media data from 24,446 products sold in a popular e-commerce platform. 3 - Does Reputation Management On Social Media Boost Career? Evidence From The Market For Executives Yanzhen Chen, University of Texas at Austin, Austin, TX, United States, yanzhen@utexas.edu, Huaxia Rui, Andrew Whinston Our paper studies the impact of reputation management (RM) on executives’ careers gathering evidence from their Twitter usage. Our structural model, which is based on a Two-Sided Matching Model, is able to exploit the characteristics of all of the candidates so as to identify separate RM’s influences in bargaining power as well as sorting during the hiring process. The results show that in the recruiting process of CEO and CMO markets, both out-/underperformed applicants benefit from RM. However, in compensation bargaining, RM is profitable only for outperformed candidates. RM can help more than 40% of CEOs and may increase their compensation by more than $10 million per year. 4 - Gravity In Open Source Software Production Xuan Ye, New York University, New York City, NY, United States, xye@stern.nyu.edu, Prasanna Tambe Using data on over 2 million hourly contributions contributed over eight years to GitHub.com, this paper investigates how the geographic organization of contributors influences open source production in corporate sponsored projects. Specifically, we test the extent to which time zones play a role in open source software contribution. We find strong evidence of such a bias in open source software production. Concentration of contributors within the same time zone is correlated with greater contribution levels, and this is robust to user fixed-effects. 211-MCC Social, Political, and Economic Applications of Social Media Analysis Invited: Social Media Analytics Invited Session Chair: Amanda Andrei, MITRE Corporation, 7515 Colshire Dr, Mclean, VA, 22102, United States, aandrei@mitre.org 1 - Characterizing Traffic Accident Detection Through Twitter And Open Data Jared Mowery, MITRE Corporation, jmowery@mitre.org Several recent studies have shown that Twitter can be used to detect traffic disruptions with high precision. This study builds on those results by incorporating open data including real-time traffic speed sensor measurements and weather data, measuring the recall of traffic accident detection against New York City police records, and by characterizing the probability of detection as a function of the severity of the accident. SC49

85