Informs Annual Meeting Phoenix 2018

INFORMS Phoenix – 2018

MB15

4 - Impact of Universal Healthcare on Patient Choice Diwas S. Kc, Emory University, 1300 Clifton Road, Goizueta Business School, Atlanta, GA, 30322, United States This paper examines the role of universal healthcare on patient choice and quality of care. Lack of health insurance is a leading contributor to Emergency Department (ED) visits from the uninsured. We find that the availability of health insurance has a significant impact on the type of healthcare sought by the previously uninsured. Specifically, individuals are less likely to continue visiting the ED, and choose hospital inpatient and outpatient services instead. We also find an effect of the policy on quality of care, as measured by same-month patient revisits. Although overall revisit rates remain unchanged, frequent users experience a decline in same-month revisits. n MB14 North Bldg 126C Operations of Contemporary Services Sponsored: Manufacturing & Service Oper Mgmt/Service Operations Sponsored Session Chair: Opher Baron, University of Toronto, Toronto, ON, M5S 1L7, Canada 1 - Matching Supply and Demand in a Service Network Levi DeValve, Duke University, 716 Turmeric Lane, Durham, NC, 27713, United States, Sasa Pekec, Yehua Wei A service provider wishes to match stochastic supply and demand located on a network. With limited resources, the service provider can only choose a subset of nodes in the network where matches can be made, and wants to choose a subset maximizing its expected profit. Although this problem is NP-hard, we provide constant factor approximation guarantees for various heuristics, including a simple greedy policy. We also discuss several applications of this problem. 2 - Spatial Pricing for Taxi Rides in New York City Nasser Barjesteh, Chicago, IL, 60615, United States, Baris Ata, Sunil Kumar We conduct an empirical study of the impact of spatial prices on the performance of the taxi industry in New York City. We use a mean-field model, in which the taxi drivers strategically search for customers in different neighborhoods across the city taking into account the distribution of the supply and demand as well as the prices across the city. We conduct a series of counterfactual analyses to explore the impact of spatial prices on the distribution of supply and demand. We also investigate how spatial prices affect the welfare of customers and drivers and how the distribution of supply, demand, the destination of the rides, and the price sensitivity of the customers impact the optimal spatial prices. 3 - Does the Bullwhip Matter Economically? A Cross-sectional Firm-level Analysis Jeffrey Callen, University of Toronto, Toronto, ON, Canada We investigate whether the bullwhip effect measured at the firm level impacts firms’ profitability. We estimate the relation between the bullwhip and various financial performance measures with a large panel of cross-sectional firm-level data. Performance is measured both in terms of mean effects and volatility effects. Our analysis yields results inconsistent with the notion that the bullwhip at the firm level has significant negative consequences on profitability. In particular, we find almost no significant statistical or economic negative relation between financial measures of profitability and the empirical bullwhip measures at the firm level, both with and without covariate controls. 4 - Learning by Doing Versus Learning by Viewing: An Empirical Study of Data Analyst Productivity at Ebay on a Collaborative Platform Yue Yin, Northwestern University, Evanston, IL, United States, Itai Gurvich, Stephanie McReynolds, Debora Seys, Jan A. Van Mieghem Effective data analytics drives business success by enhancing managerial decision- making. Companies, however, often struggle to maintain growth in the productivity of their data analysts. In this paper, we investigate how data-analyst productivity benefits from collaborative platforms that facilitate learning-by-doing (i.e. analysts learning by writing queries on their own) and learning-by-viewing (i.e. analysts learning by viewing queries written by peers). Productivity is measured using the time from creating an empty query to first executing it.

n MB15 North Bldg 127A Joint Session MSOM/Practice Curated: Empirical Service Operations Sponsored: Manufacturing & Service Oper Mgmt/Service Operations Sponsored Session Chair: Vinayak Deshpande, The University of North Carolina at Chapel Hill 1 - Improving Customer Compatibility with Operational Transparency MoonSoo Choi, Harvard Business School, 700 Soldiers Field Road, Wyss House, Boston, MA, 02163, United States, Ryan Buell Recent research has demonstrated the impact of customer compatibility - the degree of fit between the needs of customers and the capabilities of the operations serving them - on service performance. Companies with more compatible customers receive higher satisfaction scores and exhibit faster growth. However, when marketing their offerings to prospective customers, companies often shroud the operational tradeoffs inherent in their offerings in favor of emphasizing their advantages. Through a large-scale field experiment with a nationwide retail bank, we investigate how providing prospective customers with transparency into an operation’s tradeoffs affects acquisition and engagement. 2 - Don’t Call Us, We’ll Call You: An Empirical Study of Caller Behavior under a Callback Option Brett Hathaway,The University of North Carolina at Chapel Hill , 1800 Baity Hill Drive #310, Chapel Hill, NC, 27514, United States,, Seyed Emadi, Vinayak Deshpande Using call center data from a bank, we empirically study callers’ decision-making process in the presence of a callback option. We formulate a structural model of their decision-making process, and impute their underlying preferences from the data. Our estimates of their preferences show that they experience almost no discomfort while waiting for a callback, and they incur a high cost of switching from their offline tasks to answer a callback. We conduct a counterfactual analysis of how various callback policies affect the service quality and system throughput of this call center. Our results indicate that offering callbacks increases service quality without substantially impacting throughput. 3 - At Your Service on the Table: Impact of Tabletop Technology on Restaurant Performance Fangyun Tan, Southern Methodist University, 6212 Bishop Blvd, Dallas, TX, 75275, United States, Serguei Netessine We use granular data to examine the impact of a tabletop device that facilitates the order process on the check size and meal duration aspects of restaurant performance. We find that the tabletop technology is likely to improve average sales per check by 2.91% and reduce the meal duration by 9.74%, which increases the sales per minute or sales productivity by approximately 10.77%. Overall, our results indicate great potential for introducing tabletop technology in a large service industry that currently lacks digitalization. 4 - Decision Bias in the News Vendor Problem: Evidence from Airline Flight Scheduling Vinayak V. Deshpande, University of North Carolina at Chapel Hill, Kenan Flagler Business School, McColl Building, CB #3490, Chapel Hill, NC, 27514, United States, Milind Sohoni, Chandrasekhar Manchiraju Research in Behavioral Operations Management has documented “Demand Chasing” and “Pull to Center” as two prevalent behavioral biases in the single period newsvendor problem in laboratory experimental settings. Using flight- scheduling data from the US Airline industry, we show that these biases exist even in real world managerial decisions. We also show that these biases exist not only at the individual level, but are also at the firm level.

153