Informs Annual Meeting 2017

TD23

INFORMS Houston – 2017

TD23

be dramatic changes over time in customer demand and driver supply across different locations, which can result in driver shortages. One common operational strategy is to match customers with the closest driver, and to use pricing to incentivize drivers to move to undersupplied locations. This motivates a joint pricing and matching problem, and we discuss steps for its solution. 2 - Ride-hailing Networks with Platform Control Zhe Liu, Columbia Business School, New York, NY, United States, zliu18@gsb.columbia.edu, Costis Maglaras, Philipp Afeche We are motivated by the emergence of ride-hailing platforms that match demand (passengers) with service capacity (drivers) over a geographically dispersed network. Drivers are self-interested and can decide whether to join the network and whether to reposition their car when idling at a low demand location. Passenger demand is potentially unbalanced across the network in the sense that the natural passenger flow of cars need not supply sufficient capacity across network locations. The platform can exercise demand-side admission control - which passengers to match to drivers, and supply-side repositioning control - which idling drivers to move (or incentivize to do so) to high demand locations. 3 - Spatial Pricing in Ride-sharing Networks Ozan Candogan, University of Chicago, Booth School of Business, Chicago, IL, 27708, United States, ozan.candogan@chicagobooth.edu, Kostas Bimpikis, Daniela Saban We explore spatial price discrimination in the context of a ride-sharing platform that serves a network of locations. Riders are heterogeneous in terms of their destination preferences and their willingness to pay for receiving service. Drivers decide whether, when, and where to provide service so as to maximize their expected earnings, given the platform’s prices. Our findings highlight the impact of the demand pattern on the platform’s prices, profits, and the induced consumer surplus. In particular, we establish that profits and consumer surplus are maximized when the demand pattern is “balanced” across the network’s locations. 4 - Bicycle Redistribution Network Design using Crowdsourcing Jinjia Huang, Sun Yat-sen University, 135 Xingang Xi Road, Guangzhou, China, jinjiahuang@gmail.com Jinjia Huang, National University of Singapore, Singapore, Singapore, jinjiahuang@gmail.com, Mabel C.Chou, Linfeng Li, Chung-Piaw Teo Bicycle redistribution constitutes a major part of the operating cost for bicycle- sharing systems. To economize on the redistribution operations, we aim to incentivize users to take trips that can help to rebalance the network. Considering the feasibility of implementing such a system, we study whether it is enough to encourage trips on certain links of the network. We show that a system with a mix of bicycle redistribution by depot and redistribution on a small set of links suitably chosen from the full network performs almost as well as the fully flexible redistribution system using crowdsourcing. 350A Innovation/Entrepreneurship Contributed Session Chair: Ibrahim Muter, University of Bath, Bath, United Kingdom, i.muter@bath.ac.uk 1 - Investigating the Impact of Digital Business Strategy on Hospitals’ Performance Hossein Kalantar, PhD Student, Business School, University of Colorado Denver, 1475 Lawrence St., Suite 3501, Denver, CO, 80202, United States, hossein.kalantar@ucdenver.edu, Jiban Khuntia, Jahangir Karimi With the introduction of government mandates in the recent years, the adoption of Information Technology has significantly increased in healthcare organizations. However, not all healthcare organizations can fully leverage IT to enhance their performance. A clear digital business strategy can help these organizations to integrate business capabilities with new IT-based capabilities. In this study, we investigate the impact of hospitals’ digital business strategy on their performance. For this study, we conduct hospitals executive survey and combine it with secondary data from the American Hospital Association’s (AHA) and the Healthcare Information and Management System Society (HIMSS). 2 - Can We Trust Online Physician Ratings? Evidence from Cardiac Surgeons in Florida Susan F. Lu, Purdue University, Krannert 441, West Lafayette, IN, 47907, United States, lu428@purdue.edu, Huaxia Rui Despite heated debate about the pros and cons of online physician ratings, little systematic work has examined the correlation between physicians’ online ratings and their actual medical quality. Using the ratings of cardiac surgeons at RateMDs and the patient outcomes of coronary artery bypass graft surgeries in the 2013 Florida Hospital Inpatient Discharge Data, we investigate whether online ratings are informative about physicians’ medical quality. TD25

342E Revenue Management with Consumer Analytics Sponsored: Revenue Management & Pricing Sponsored Session Chair: Joline Uichanco, University of Michigan, Ross School of Business, Ann Arbor, MI, 48109-1234, United States, jolineu@umich.edu 1 - Data-driven Pricing for a New Product Mengzhenyu Zhang, University of Michigan, Ross School of Business, Ann Arbor, MI, 48105, United States, zhenyuzh@umich.edu, Hyun-Soo Ahn, Joline Uichanco Due to the lack of relevant data, decisions regarding new products are difficult to make, and any mistake has grave consequences to the firm’s bottom line. The famous Bass model for new product adoptions requires a priori knowledge on certain parameters which can be determined only after collecting demand data. We study a problem of pricing for a new product under stochastic Bass model when market parameters are uncertain. We propose simple and efficient data- driven pricing approaches and characterize their performances. Our framework is sufficiently flexible to be applied to real time parameter learning in other general diffusion processes along with control reoptimization. 2 - Value of Delayed Incentive: An Empirical Investigation of Gift Card Promotions Bharadwaj Kadiyala, Univ at Texas-Dallas, 2400 Waterview Parkway, Apt 831, Richardson, TX, 75080, United States, bxk121930@utdallas.edu, Ozalp Ozer, Serdar Simsek Gift cards have become a popular vehicle for promotional campaigns run by many departmental, consumer electronic, and online retail stores. Using a proprietary data set from a large department store, we investigate how customers respond to these promotions as well as its effectiveness as a promotional vehicle for retailers. 3 - Customer-trends for Personalized Demand Estimation and Targeted Promotions Kiran Venkata Panchamgam, Oracle America Inc., 78 Kendall Ct, Bedford, MA, 01730, United States, kpanch@gmail.com, Georgia Perakis, Setareh Borjian, Lennart Baardman, Tamar Cohen Detecting trends in fashion can help retailers determine effective personalized promotion plans more easily. Access to social media data can be important in order to be able to understand these trends. Unfortunately, social data is usually unavailable. We introduce a personalized demand model that captures customer trends and show an efficient method for estimating it from transaction data. Last, we show that the promotion targeting problem is NP-hard to solve exactly, and develop an efficient greedy approach that scales to large datasets and finds provably near-optimal solutions. 4 - A Dynamic Clustering Approach to Data-driven Assortment Personalization Sajad Modaresi, Duke University, Durham, NC, United States, sajad.modaresi@duke.edu, Fernando Bernstein, Denis R. Saure A retailer faces heterogeneous customers with initially unknown preferences. The retailer can personalize assortment offerings based on available profile information; however, users with different profiles may have similar preferences, suggesting that the retailer can benefit from pooling information. We propose a dynamic clustering policy that adaptively adjusts customer segments and personalizes the assortment offerings. We test the policy’s performance using a dataset from a Chilean retailer. 342F Ride-sharing Platforms I Sponsored: Revenue Management & Pricing Sponsored Session Chair: Ozan Candogan, University of Chicago, Chicago, IL, 27708, United States, ozan.candogan@chicagobooth.edu Co-Chair: Daniela Saban, Stanford GSB, Palo Alto, CA, 94304, United States, dsaban@stanford.edu 1 - On Pricing and Matching in Ridesharing Systems Erhun Ozkan, University of Southern California, 3670 Trousdale Pkwy, Bridge Hall, 401, Los Angeles, CA, 90089, United States, erhunozkan@gmail.com, Ramandeep Randhawa, Amy Ward Ridesharing platforms are two-sided matching markets that pair customers and drivers. The ridesharing company would like to ensure there is always a nearby driver to offer an arriving customer. The difficulty is that drivers choose when to work, how long to work, and where to go to search for customers. The result can TD24

384