2016 INFORMS Annual Meeting Program

TD36

INFORMS Nashville – 2016

TD35 205A-MCC On Demand Services Sponsored: Manufacturing & Service Oper Mgmt, Service Operations Sponsored Session Chair: Pnina Feldman, University of California-Berkeley, Haas School of Business, Berkeley, CA, 94720, United States, feldman@haas.berkeley.edu Co-Chair: Robert Swinney, Duke University, Fuqua Drive, Durham, NC, 27708, United States, robert.swinney@duke.edu 1 - Drivers, Riders And Service Providers: The Impact Of The Sharing Economy On Mobility Harald Bernhard, Singapore University of Technology and Design, Singapore, Singapore, harald_bernhard@mymail.sutd.edu.sg, Saif Benjaafar, Costas Courcoubetis We study a heterogeneous population of agents interacting through a platform that facilitates on-demand ride-sharing. We build an equilibrium model to analyze the impact of key parameters such as car usage and ownership costs on traffic volume and welfare. Furthermore we define and find conditions to differentiate between a ‘need’ and ‘profit’ driven sharing economy. 2 - The Role Of Surge Pricing On A Service Platform With Self- scheduling Capacity Gerard P Cachon, University of Pennsylvania, cachon@wharton.upenn.edu, Kaitlin Daniels, Ruben Lobel Recent platforms, like Uber and Lyft, offer service to consumers via “self- scheduling” providers who decide for themselves how often to work. These platforms may charge consumers prices and pay providers wages that both adjust based on prevailing demand conditions. We study the effectiveness of different contractual forms, from the perspective of platform profit, provider surplus and consumer surplus. We find that while surge pricing is not optimal, it is nearly so. We describe conditions under which all parties benefit from the use of surge pricing. 3 - Bike-share Systems: Accessibility And Availability Ashish Kabra, INSEAD, Boulevard de constance, Fontainebleau, 77305, France, ashish.kabra@insead.edu, Elena Belavina, Karan Girotra This paper estimates the relationship between ridership of a bike-share system and its design aspects— station accessibility and bike-availability. Our analysis is based on a structural demand model that considers the random-utility maximizing choices of spatially distributed users, and it is estimated using high- frequency system-use data from the bike-share system in Paris and highly granular data on sources of bike-share demand. A novel model separates the long-term and short-term effects of higher bike-availability. Because the scale of our data render traditional numerical estimation techniques infeasible, we develop a novel transformation of our estimation problem. TD36 205B-MCC MSOM/Supply and Procurement Sponsored: Manufacturing & Service Oper Mgmt, Supply Chain Sponsored Session Chair: Zhixi Wan, University of Oregon, 1208 University of Oregon, Eugene, OR, 97403, United States, zwan@uoregon.edu 1 - Optimal Procurement In Assembly Supply Chains Bin Hu, University of North Carolina, Chapel Hill, NC, 27519, United States, bin_hu@unc.edu, Anyan Qi We consider an OEM’s contracting mechanism to procure multiple components from different suppliers and assemble them into products under simultaneous and sequential contracting. We derive optimal mechanisms in both cases, and show that they can be implemented by simple quantity flexibility contracts. Furthermore, we find that optimal simultaneous and sequential contracting are revenue-equivalent for all parties, despite them having different asymmetric information structures. All results are extended to general convex costs and concave revenues, confirming that the results capture fundamental properties of optimal procurement in assembly supply chains.

3 - Queue Now Or Queue Later Brett Hathaway, Doctoral Candidate, UNC Chapel Hill, Chapel Hill, NC, United States, Brett_Hathaway@kenan-flagler.unc.edu, Seyed Emadi, Vinayak Deshpande We study caller redial behaviors using call center data from a US-based bank. We show which factors affect the probability of redialing and the time between queue abandonment and redial. Using structural estimation, we show through counterfactual experiments how a center with callers who redial performs under various routing policies. 4 - Access Times In Appointment-driven Systems And Level- dependent MAP/G/1 Queues Petra Vis, VU Amsterdam, De Boelelaan 1105, Amsterdam, 1081 HV, Netherlands, petra.vis@vu.nl Petra Vis, Centrum Wiskunde & Informatica, Amsterdam, Netherlands, petra.vis@vu.nl, Rene Bekker We study access times in appointment-driven systems. The access time is the number of days between a request for an appointment and the day that the appointment can take place. To meet target access times, we allow for overbookings, as they often occur in practice. Applications of this type of systems can be found in health care; e.g., patients making appointments with a doctor. We argue that such a system can naturally be modelled as an MAP/G/1 queue; the level corresponds to the access time and the phase to the dynamics of the number of free slots at the first available day. To allow for overbookings, we analyze a level-dependent version of the MAP/G/1 queue, leading to intuitively appealing results. 204-MCC Joint Session HAS/MSOM-HC: Care Transition Policy and Management Sponsored: Manufacturing & Service Oper Mgmt, Healthcare Operations Sponsored Session Chair: Nan Kong, Purdue University, West Lafayette, IN, United States, nkong@purdue.edu 1 - System Modeling For Patient Transitions Within Hospital Hyo Kyung Lee, University of Wisconsin Madison, 1, Madison, WI, 1, United States, wilchess27@gmail.com, Jingshan Li, Albert J. Musa, Philip A. Bain Among various issues in healthcare delivery, many complex and critical problems occur at the interfaces of healthcare systems. A patient’s hospital stay may encompass various care units, but due to limited capacity, substantial amount of patients experience delay during the transition. This not only impacts the care quality and patient satisfaction, but in some cases is directly associated with mortality risk. Thus, to contribute to this end, we present a Markov chain model to study the transitions between emergency department, intensive or critical care unit, and hospital ward in community hospitals. Furthermore, an iteration method is introduced to evaluate the performance. 2 - Reduce COPD Readmission - Risk Identification And Patient- centered Intervention Xiang Zhong, University of Wisconsin, 1513 University Avenue, Room 3235, Madison, WI, 53706, United States, oliver040525@gmail.com, Cong Zhao, Philip Bain, Albert Musa, Craig Sommers 30-day hospital readmission has been established as a critical performance indicator in promoting quality and patient-centered care. Individuals with serious chronic conditions such as chronic obstructive pulmonary disease (COPD) suffer high readmission risks and incur significant hospital penalty cost. To reduce COPD readmission, it’s important to provide tools for physicians and hospitals to manage patients post discharge. In this study, we build statistical models to identify the risk factors for COPD readmission. Based on patients’ risk levels, different patient- centered intervention policies prior to discharge and post discharge are developed. 3 - Optimal Inpatient Discharge Planning Under Uncertainty Maryam Khatami, Texas A&M University, 4050 ETB, College Station, TX, 77840, United States, maryam.khatami@tamu.edu, Mark Lawley, Nan Kong, Michelle M. Alvarado We study the inpatient discharge planning problem to enable efficient design of optimal discharge plans on a daily basis. If some of the discharge processes are delayed, the ensuing backup in the upstream units will cause inpatient admission delays. Hence, it is critical to tradeoff competing issues of upstream patient boarding (e.g. Emergency Department (ED) boarding), inpatient discharge lateness, and Inpatient Unit (IU) workload integration. We develop a novel two- stage stochastic programming model with uncertain IU discharge processing time and IU bed request time. Using data from a Texas hospital, we calibrate our model and fine-tune our solution method. TD34

345