Informs Annual Meeting Phoenix 2018

INFORMS Phoenix – 2018

SC57

Stop B6000, Austin, TX, 78712, United States The annual supply of human organs in the US has been more or less stable in recent years. The demand has outstripped supply and the national waitlist continues to increase every year. This supply-demand imbalance leads to human suffering, failure to access transplantation, and waitlist deaths. Many Operations Researchers have focused attention on the national system of matching and prioritizing potential transplant recipients, which affects organ utilization. Some other works have focused on geographic disparities between supply and demand and suggested improvements via redesign. These efforts focus only on a subset of practices at both procurement and utilization ends of organ supply that result in reduced supply. This talk will describe current practices across the organ procurement and transplantation system that may benefit from OR models, present preliminary models, and suggest research opportunities. n SC57 West Bldg 101B HAS Session Sponsored: Health Applications Sponsored Session Chair: Jayashankar M. Swaminathan, University of North Carolina Chapel Hill, Chapel Hill, NC, 27599-3490, United States Co-Chair: Sandeep Rath 1 - Financial Incentives Under CPC+ Elodie Adida, University of California at Riverside, CA, United States, Fernanda Bravo The Centers for Medicare and Medicaid Services have recently launched the Comprehensive Primary Care Plus (CPC+) initiative, aiming at improving primary care delivery by changing the way providers are paid for their services. CPC+ includes a capitation payment, a reduced fee per visit, as well as a performance- based payment incentive. Under this program, physicians are encouraged to use alternative care delivery methods (phone calls, e-visit, in-home nurse visits, etc.). We study how this payment system impacts providers’ care delivery decisions, patient welfare and payer cost. 2 - Sustainability Planning for Healthcare Information Exchanges Tharanga Rajapakshe, Chelliah Sriskandarajah, Subodha Kumar, Arun Sen We investigate a multi-period two-service model where a HIE offers (i) the healthcare information sharing service, and (ii) the supplier rebate program. Our proposed model is primarily based on our interactions with a number of HIE providers in Texas. First, we present structural properties and equilibrium solution for our model. Then, based on extensive computational experiments, we present several useful managerial insights for the HIE provider as well as the policy- makers. 3 - Coordinated Care for Mental and Physical Health Sandeep Rath, University of North Carolina at Chapel Hill - Kenan Flagler, CB #3490, McColl Building, McColl 4705, Chapel Hill, NC, 27599, United State, Jayashankar M. Swaminathan Multiple clinical trials have demonstrated the benefits of coordinating treatment of mental health conditions like depression and chronic physical conditions like diabetes. However, sustainability of coordinated care outside trial settings has not been fully demonstrated. A sustainable coordinated care will be revenue neutral for the providers and improve patient outcomes. Towards this, we propose a mathematical optimization model which would optimize care management plans to improve patient outcome while balancing revenue and resource usage costs. 4 - Estimating Patient Flow Models via Robust Queueing Theory Chaithanya Bandi, Kellogg School of Management, Northwestern University, 2211 Campus Dr, room 4169, Evanston, IL, 60208, United States In this paper, we consider the statistical study of partially observed queueing systems arising in application areas such as Hospital networks, data centers, and cloud computing systems. Because these services operate under strict performance requirements, a statistical understanding of their performance is of great practical interest. A key challenge in these settings is that the data are incomplete, because recording detailed information about every request to a heavily used system can require unacceptable overhead. We propose an analytically tractable approach for studying inference problems in these queueing systems. Our approach is based on the Robust Queueing framework.

n SC55 North Bldg 232C Joint Session Practice Curated/Practice Sports Analytics II Sponsored: SpORts Sponsored Session Chair: Stephen Hill, UNC Wilmington, United States 1 - NFL versus NCAA Football Decision Making Analysis Stephen Hill, UNC Wilmington, United States In this work in game win probability models are developed for NFL and NCAA football. These models are compared and then used to evaluate decision making approaches. Additionally, consideration is given to measuring in game win probability volatility. 2 - Real-Time NBA Playoff Elimination Mark Husted, Colorado School of Mines, 816 20th Street, Golden, CO, 80401, United States, Alexandra M. Newman, Eli Olinick The NBA is divided into two conferences, each of which is comprised of fifteen teams, and the top eight teams from each conference compete in the playoffs. An integer-programming model determines when a team has been eliminated from the playoffs before the completion of the regular season. There are instances in which teams’ winning percentages are tied. Ties are broken using seven independent criteria based on the number of teams tied. The results are published on the RIOT website so fans can follow their favorite teams’ playoff standings. We compare the time at which (and day on which) these results are published to the NBA official standings; in many cases, RIOT notifies the public prior to the NBA. 3 - Learning from Success and Failure at the Speed of Formula One Michael A. Lapré, Vanderbilt University, 401 21st Avenue South, Nashville, TN, 37203, United States, Candace Cravey KC, Staats and Gino (2013) found that surgeons learn from their own success and from others’ failure. Unlike surgery, in Formula One racing, success is rare and competitive. Using data since Formula One started in 1950, we investigate driver learning from own experience and teammates’ experience with both success and failure. We find that drivers learn from own success, teammates’ success, as well as own car failures. We use characteristics of success to explain observed learning effects. 4 - Maximizing National Hockey League Goal Differential through the Examination of Line Orders and Lengths William Davis, Student, Slippery Rock University, Slippery Rock, PA, 16057, United States, Jacob Lindey While graph theory and mathematical predictive measures rarely enter into dialogue with hockey, these are the very concepts that would provide National Hockey League managers with better chances of winning through more efficient ice time. We examine and implement these concepts to manipulate line orders and lengths to maximize goal differential for any given team. In addition, we break down our easy-to-use interface designed for managers. 5 - A Classification and XML Framework for Round Robin Sports Timetabling Problems Dries Goossens, Ghent University, Tweekerkenstraat 2, Gent, 9000, Belgium, David Van Bulck, Jorn Schonberger, Mario Guajardo In the sports timetabling literature, most problems assume different constraints and objectives. Often, one specific case study is discussed, with an accompanying solution method. This lack of structure makes it hard to compare problem instances and to assess algorithmic performance. We present a 3-field classification scheme for round-robin sports timetabling problems. For the instance definition and its solution(s), we propose two XML-file-based templates and a C++-library (RobinX) to evaluate both files. We present a web application that should encourage exchanging instances and solutions, and eventually lead to more general insights. n SC56 West Bldg 101A HAS Distinguished Scholar Lecture Diwakar Gupta Sponsored: Health Applications Sponsored Session Chair: Ebru Korular Bish, Virginia Tech, Blacksburg, VA, 24060, United States Co-Chair: Tinglong Dai, Johns Hopkins University, Baltimore, MD, 21202, United States 1 - The US Deceased-donor Organ Procurement and Utilization System Diwakar Gupta, Professor, University of Texas, 2110 Speedway

81