Informs Annual Meeting Phoenix 2018

INFORMS Phoenix – 2018

MA14

1 - Pooling Queues in Discretionary Services Guillaume Roels, INSEAD, Boulevard de Constance, Fontainebleau, 77305, France, Mor Armony, Hummy Song Contrary to the classical theory of operations management, recent case studies in retail, call centers, and healthcare indicate that pooling queues may not necessarily result in less expected work in process. In this paper, we propose that this phenomenon may arise when servers choose their own capacity to trade off their capacity costs with the holding costs of customers in service (when the queue is not visible to the servers) or of all customers in the system (when the queue is visible). We show that the difference in operational performance between the two configurations is marginal when the queue is visible, but can be substantial, with a preference for dedicated queues, when the queue is not visible. 2 - Service Delivery Platforms: Pricing, Welfare, and Revenue Implications Andrew E. Frazelle, Duke University-Fuqua School of Business, 540 South LaSalle Street, #4340, Durham, NC, 27705, United States, Pnina Feldman, Robert Swinney Service delivery platforms maintain a symbiotic relationship with the existing providers in their industry. While this relationship entails cooperation between the platform and the restaurant, there is substantial opportunity for misalignment. We model the restaurant’s kitchen as an M/M/1 queue with customer waiting costs. We first study the revenue maximization problem faced by a monopolist who controls both the dine-in and delivery prices and receives all revenues from the system. We then investigate means of coordinating this supply chain via different contracts between the restaurant and the platform, and we find that a two-way revenue-sharing contract coordinates the system. 3 - Managing the Interaction of Acquisition and Retention in Customer Intensive Services Vasiliki Kostami, London Business School, Regent’s Park, London, NW1 4SA, United Kingdom, Vasiliki Kostami, HEC Paris, Paris, France, Oded Koenigsberg Customer acquisition affects customer retention and vice versa due to budget constraints but this is not the sole factor. Loyal customers and potential customers have different reception of prices and the quality they are offered which adds an extra layer of interaction between the two segments. We study the optimal prices and service rates in customer intensive services under the consideration of the linking between customer acquisition and retention. 4 - Overbooking with Endogenous Demand Jingxing (Rowena) Gan, Wharton School of Business, University of Pennsylvania, Philadelphia, PA, 19104, United States, Gerry Tsoukalas, Noah Gans Traditional models of airline overbooking assume exogenous demand that is not affected by the booking policy. We study how the overbooking policy can affect consumer demand ex ante. Given a rational choice model for consumers, we derive an airline’s optimal booking, pricing and compensation policy for a single flight with a deterministic individual no-show rate. We model and compare two compensation schemes, a fixed scheme and an auction scheme. Our results show that the airline overbooks less when considering strategic consumer behavior as opposed to treating demand as exogenous. n MA16 North Bldg 127B Operations Management and Development Sponsored: Manufacturing & Service Oper Mgmt/Sustainable Operations Sponsored Session Chair: Andre Du Pin Calmon, INSEAD, Fontainebleau, 77300, France 1 - Optimizing First-mile Logistics in Smallholder Supply Chains Sergio Camelo, Stanford University, Stanford, CA, United States, Joann de Zegher, Dan Andrei Iancu, Daniela Saban We study the logistics surrounding pick-ups and deliveries of palm fruit in the Indonesian smallholder palm oil supply chain. Using data from roughly 3,000 smallholder farmers and 100 transport providers from our field site in Sumatra, we reconstruct current transportation routes, and propose novel algorithms that rely on spatial or temporal optimization. We find that centralized vehicle routing algorithms can improve the transportation distances by 25%, and reduce the number of days that middlemen have to work by 30%. We expect the implementation of our results to reduce overall costs in the market, and ease the digitization of agricultural data in smallholder palm oil supply chains.

n MA14 North Bldg 126C

Innovation in Service Management and Delivery Sponsored: Manufacturing & Service Oper Mgmt/Service Operations Sponsored Session Chair: Serguei Netessine, The Wharton School, Philadelphia, PA, 19104, United States 1 - Investing in Performance: Information and Merit-based Incentives in K-12 Education Vanitha Virudachalam, Wharton School, OID, 3730 Walnut Street, 500 Jon M. Huntsman Hall, Philadelphia, PA, 19104, United States, Sergei Savin, Matthew Steinberg We study performance-based contracts in a K-12 educational setting, where a school district contracts with teachers to meet student performance goals. We examine the trade-offs between information and monetary incentives faced by districts using a two-period dynamic principal-agent model, where the school district (principal) decides on the level of merit-based incentive to offer and whether to invest in assessments that provide more accurate mid-year student performance information and the teachers (agents) decide the level of effort to exert in each period. We identify settings where the benefits of the additional information justify their costs. 2 - When to Innovate and when to Grow? Empirical Analysis of Startup Operation Christophe Pennetier, INSEAD, 6 Marina Boulevard, # 27-15, Singapore, 018985, Singapore, Serguei Netessine, Karan Girotra Using a new curated dataset with more than half a million startups, we use customized text mining / machine learning techniques to identify business model changes and study their effects on startups’ success. We are the first to empirically relate startup operations to performance. 3 - Efficiency of Streethail vs. Booking Apps Jussi Keppo, National University of Singapore, Mochtar Riady Building, BIZ 1 8-69, 15 Kent Ridge Drive, Singapore, 119245, Singapore, Shih-Fen Cheng, Ming Hu We study the efficiency of taxi divers to serve passengers by using streethail vs. booking apps. Our key metric is the percentage of time the taxi drivers spend in serving their passengers. We find that on average the service time of streethail and booking is about the same. Although booking technologies reduce the roaming time, the resulting response time makes the booking as efficient as streethail. We study the heterogeneity of our findings, and show that guided roaming combines the advantages of streethail and booking apps, and improves the efficiency of taxi drivers. 4 - Reading Between the Stars: Understanding the Effect of Online Customer Reviews on Product Demand Many studies have examined quantitative customer reviews (i.e. star ratings) and found that they are a relevant source of quality information. We also examine the qualitative part of customer reviews by implementing a novel algorithm to assess the sentiment of the text and study how the interplay between sentiment and star ratings relates to product demand. Using the U.S. Automobile market data, we find that sentiment and star ratings both have a decreasingly positive effect on demand and that their interaction effect suggests they act as non-trivial complements: positive sentiments in text reviews compensate for the tendency to discount highly positive ratings. n MA15 North Bldg 127A Service Systems with Strategic Behavior Sponsored: Manufacturing & Service Oper Mgmt/Service Operations Sponsored Session Chair: Noah Gans, University of Pennsylvania, Philadelphia, PA, 19104- 6340, United States Co-Chair: Gerry Tsoukalas, Wharton School of Business,Philadelphia, PA, 19104, United States Co-Chair: Rowena Gan, Wharton School of Business, Philadelphia, PA, 19104, United States Hallie Cho, INSEAD, Boulevard de Constance, PhD Office, Fontainebleau, 77300, France, Sameer Hasija, Manuel Sosa

124