Informs Annual Meeting 2017

TB10

INFORMS Houston – 2017

TB10

2 - Listen to the Crowd: Network Effects and Online Reviews in Restaurant Sales Forecasting Shawn Mankad, Cornell University, Ithaca, NY, United States, spm263@cornell.edu, Qiuping Yu, Masha Shunko Using a comprehensive dataset from a major restaurant franchise, we forecast weekly store sales using classical measures of service quality from internal surveys at focal and neighboring stores, in addition to online ratings. Our results show that higher quality at the neighboring stores leads to higher sales at the focal store. We also find that the accumulation of online reviews reduces the importance of internal quality surveys at focal and neighboring stores as predictors. 3 - The Customer May Not Always be Right: Customer Compatibility and Service Performance Ryan Buell, Harvard Business School, Morgan Hall 429, Boston, MA, 02163, United States, rbuell@hbs.edu, Dennis Campbell, Frances X. Frei Leveraging data from a nationwide retail bank, we investigate the extent to which customer satisfaction is explained by differences between customers (versus differences among factors more directly controlled by the firm). We find that differences among customers account for more than 90% of the explainable variation in customer satisfaction, and that although individual customers tend to experience the operation consistently, there are surprisingly large differences in experiences among customers. The presentation will also preview ongoing research that highlights how operational design choices can shape customer selection, and in turn, customer compatibility. 4 - Ratings and Version Updates in the Mobile App Market Gad Allon, University of Pennsylvania, gadallon@wharton.upenn.edu, Yorgos Askalidis, Randall Berry, Nicole Immorlica, Ken Moon, Amandeep Singh Mobile apps have become an economy with a projected market size of $77 Billion in 2017. One of the key features that distinguishes mobile apps from other types of digital goods (such as movies, songs or books) is that they have versions. A developer can release an app into a mobile app store, and can then keep adding, removing, or editing features of the app with subsequent version updates. We characterize the optimal strategy for timing such updates and structurally estimate the extent to which developers respond to demand-side incentives in a ratings- sensitive environment. We discuss implications for platforms and developers. 332C Game Theory Contributed Session Chair: Jian Yang, Rutgers University, Newark, NJ, United States, jyang@business.rutgers.edu 1 - Multi-strategy Trust Evolution Dynamics in the Venture Capital Syndicate Network Heyin Hou, Southeast University, School of Economics and Management, Southeast University, Nanjing, 211189, China, heyinhou@hotmail.com, Jun Yin, Weida Chen Venture Capital Syndicate Network (VCSN) derives from syndicate investments and relies on trust between venture capitalists (VCs). Based on a Wright-Fisher process, we introduce a general game model of trust evolution dynamics with multi-strategy and protection mechanism in the VCSN. Then, we explore influences of the protection mechanism on evolutionary stability conditions for the two-strategy and the four-strategy game models. Finally, we construct the global profit function for the VCs and adopt simulation experiments to verify the trust evolution dynamics for the two-strategy and the four-strategy game models. 2 - Augmented Probability Simulation Methods for Non-cooperative Games Tahir Ekin, Assistant Professor of Quantitative Methods, Texas State University, 601 University Dr. Mccoy 451, San Marcos, TX, 78666, United States, t_e18@txstate.edu We provide an augmented probability simulation framework to solve non- cooperative games, focusing on sequential problems. We include approaches to approximate Nash equilibria under common knowledge conditions, assess such solutions under a robust approach and, finally, approximate adversarial risk analysis solutions when lacking common knowledge. A security example serves as illustration. TB13

330B Data and Information in Operations Sponsored: Manufacturing & Service Oper Mgmt Sponsored Session

Chair: Fuqiang Zhang, Washington University in St. Louis, St. Louis, MO, 63130, United States, zhang@olin.wustl.edu 1 - Trade-off Between Flexibility and Productivity: An Empirical Analysis of Manufacturing Networks Vivek Choudhary, INSEAD, Singapore, Singapore, Vivek.Choudhary@insead.edu, Sameer Hasija, Serguei Netessine Manufacturing flexibility has been studied in many contexts and the celebrated “chaining” principle dictates how manufacturing networks should be structured. However, empirical analysis of this principle are non-existent. We analyze the impact of manufacturing network structure on productivity. 2 - Consumer Equilibrium, Pricing, and Efficiency in Group Buying: Theory and Evidence Tunay Tunca, Unversity of Maryland, 8803 Courts Way, College Park, MD, 20742, United States, ttunca@rhsmith.umd.edu, Liu Ming We study consumer equilibrium and network effects in group buying mechanisms. We theoretically model dynamic consumer behavior in group buying and derive the continuous time stochastic recursive equilibrium. Utilizing data from a major retail platform, we structurally estimate the model and predict Zekun Liu, Washington University in St. Louis, One Brookings Drive, Campus Box 1156, St. Louis, MO, 63130, United States, liu.zekun@wustl.edu, Fuqiang Zhang, Dennis Zhang We study the information sharing strategy for a platform on which independent sellers engage in Cournot competitions. The platform possesses superior information than the sellers and may choose to share the information with the sellers. We characterize the optimal sharing strategy for the platform. 4 - Corporate Social Responsibility with Asymmetric Information Xiaomeng Guo, Hong Kong Polytechnic University, Hong Kong, Hong Kong, xiaomeng.guo@polyu.edu.hk, Guang Xiao, Fuqiang Zhang We develop a game-theoretic model to analyse the impact of consumers’ social- responsibility concerns on a firm’s pricing decision and profit. The firm can be either socially responsible or non-responsible, and consumers do not have perfect information about the firm’s type. We find that the existence of information asymmetry may have interesting implications for the firm regarding corporate social responsibility. 332A Empirical Service Operations Sponsored: Manufacturing & Service Oper Mgmt, Service Operations Sponsored Session Chair: Pnina Feldman, University of California-Berkeley, University of California-Berkeley, Berkeley, CA, 94720, United States, feldman@haas.berkeley.edu 1 - Welfare Implications of Congestion Pricing: Evidence from SF Park Hsin-Tien Tsai, University of California, Berkeley, CA, 94703, United States, hsintien@berkeley.edu Congestion pricing offers an appealing solution to urban parking congestion. We find that SFpark increases consumer and social welfare in congested regions but may hurt welfare in uncongested regions. Interestingly, congestion pricing may not surely reduce search traffic, because dispersed prices induce consumers to search for affordable spaces. In such cases, a simpler pricing policy may achieve higher welfare than a complex one. Lastly, congestion pricing increases social welfare compared to capacity rationing. The insights offer important implications for local governments considering regulation vs. market-based approaches for managing parking. TB11 the efficiency gains created by employing the mechanism. 3 - Demand Information Sharing for a Platform

306