2016 INFORMS Annual Meeting Program

WA50

INFORMS Nashville – 2016

WA50

3 - The Future Burden Of CKD In China: A Simulation Model For The CKD Initiative Nan Chen, Tsinghua University, Room 615, Shunde Building, Tsinghua University, Haidian District, Beijing, 100084, China, chenn618@gmail.com, Jinwei Wang, Xiaolei Xie, Luxia Zhang, Li Zheng The prevalence of chronic kidney disease (CKD) is high in China, which is approximately 10.8% in 2010. However, awareness of CKD remains low, only 12.5% of the 119.5 million patients are aware of the condition. There exist very few studies to estimate the future burden of CKD. We developed a CKD Health Policy Model for Chinese people based on annual decrements in estimated glomerular filtration rates that depend on age and risk factors. We used this model to simulate the residual lifetime incidence of CKD and project the prevalence of CKD in China.

212-MCC SpORts: Sports Analytics III

Sponsored: SpORts Sponsored Session

Chair: Scott Nestler, University of Notre Dame, Mendoza College of Business, Notre Dame, IN, 46556, United States, snestler@nd.edu 1 - National Hockey League Goaltending: An Analysis Of Goaltender Performance In Relation To Amount Of Days Rest Paul Weisgarber, U.S. Air Force Academy, USAF Academy, CO, United States, paul.a.weisgarber@gmail.com, Jeremy Forbes, Luke Guinan When deciding which goalie to start, professional hockey teams have historically made that decision based on who the better overall goaltender is and whether they need a night (or more) rest. Aside from coach and player intuition, little data has been involved in such decisions. Motivated by an SB Nation article on broadstreethockey.com, we attempt to better inform NHL coaches and general managers on the relationship between the number of days rest between games (DRBG) for a goalie, his save percentage (SV%), and team wins and losses. 2 - Analysis Of Corner Kicks In Football (Soccer) Nils Rudi, INSEAD, Nils.Rudi@insead.edu, Tong Wang Using coded events and tracking data from football matches in a major football (soccer) league, we (1) study the prediction of the number of corner kicks in a match statically (using only the information available before the match starts) and dynamically (using live feed of critical events) and (2) investigate the dynamics that convert an awarded corner kick into a goal and factors that affect the conversion rate. 3 - Is Strength Of Schedule A Real Strength For NFL Teams Ismail Civelek, WKU, ismail.civelek@wku.edu, Murat Kurt The National Football League (NFL) uses both complex analytical tools and panel of experts to schedule regular season games to assure owners, coaches, players and fans that no team has an advantage. The strength of schedule has been a major disagreement in scheduling NFL games due to ongoing dispute about this measure. This paper proposes a mixed-integer-linear program to investigate the relationship between the strength of schedule and teams’ making into the play-off and tries to answer whether the strength of schedule is a real strength for the NFL teams. 213-MCC Evaluating Health Systems of Public Interest Sponsored: Public Sector OR Sponsored Session Chair: Andres Garcia-Arce, University of South Florida, USF, Tampa, FL, 3, United States, andresg@mail.usf.edu 1 - Hospital Preventable Readmissions And Interventions In Medicare Patients Andres Garcia-Arce, University of South Florida, Tampa, FL, United States, andresg@mail.usf.edu, Jose L. Zayas-Castro Hospital preventable readmissions in the US are considered as a target for quality improvement by the affordable care act. Medicare uses economic penalties for hospitals with excessive readmissions. National experts present concerns about the appropriateness and fairness of these measurements such as the excessive impact on safety net hospitals. The use of disease-specific interventions reduces readmissions while directly improves the quality of care and produce savings. This research aims to use disease-specific health interventions to reduce readmissions. The results from this work are intended to open a discussion on alternative policies to address preventable readmissions. 2 - Predicting Likelihood Of Drug Approval From Clinical Trials Felipe A Feijoo, Johns Hopkins University, ffeijoo@jhu.edu, Sauleh Ahmad Siddiqui, Jenny Bernstein Pharmaceutical companies face huge risks and costs in order to launch a new drug to market. These costs are associated with expensive and timely clinical trials with a success rate that from 10% to 20%. In order to understand the drivers that make drugs to fail at some stage of a clinical trial, we developed a machine learning (based random forest) to determining the factors that are associated with clinical success. Our model is capable to predict with an 85% accuracy the new compounds that will get FDA marketing approval. WA51

WA52

214-MCC Network Repair and Resiliency for Service Restoration Sponsored: Public Sector OR Sponsored Session

Chair: Ozlem Ergun, Northeastern University, 453 Meserve, 360 Huntington Avenue, Boston, MA, 02115, United States, o.ergun@neu.edu Co-Chair: Aybike Ulusan, Northeastern University, 360 Huntington Avenue, Boston, MA, 02115, United States, ulusan.a@husky.neu.edu 1 - Network Science Based Quantification Of Resilience Of Multi- scale Infrastructure Systems Udit Bhatia, Northeastern University, bhatia.u@husky.neu.edu Natural or human-induced disruptions to multi-scale critical lifeline infrastructure networks can damage economies and cause loss of lives. Characterizing brittleness and guiding restoration are crucial for post-hazards recovery and proactive design. Here we develop a quantitative network-science framework to understand fragility and resilience of interdependent lifelines, which we demonstrate on the interdependent Boston Mass Transit, Power transmission system by assessing robustness and evaluating recovery strategies. Natural hazards and cyber-physical attacks, as well as non-systematic and cascading infrastructure failures are considered. 2 - Transportation Network Recovery Based On Multi-industry Economic Impact Mohamad Darayi, The University of Oklahoma, mdarayi@ou.edu, Kash Barker, Nazanin Morshedlou Freight transportation networks, considered a means to enable the flow of commodities and to facilitate economic productivity, are prone to natural and human-made hazards. This research pursues an approach to improve restoration order decision making based on the broader perspective of their impact to multiple industries and multiple regions. 3 - On The Cost Of Decentralized Scheduling For Interdependent Network Restoration Hongtan Sun, Rensselaer Polytechnic Institute, 110 8th St., Troy, NY, 12180, United States, sunh6@rpi.edu, Thomas Sharkey We consider the problem of restoring disrupted services across multiple interdependent networks after extreme events. The restoration efforts are usually formulated in a decentralized manner as each system optimizes their own restoration schedule. We consider integer programming approaches to determine the equilibrium (stable) solutions for this decentralized scheduling system. These approaches help to calculate the price of anarchy and the price of stability which help to measure the loss in the centralized objective from the decentralized scheduling process. 4 - Restoration Of Network Connectivity In Large-scale Disaster Management Problems Aybike Ulusan, Northeastern University, ulusan.a@husky.neu.edu, Ozlem Ergun The goal of this study is to offer enlightening insights on the network restoration problems by developing network science based quantitative frameworks. As the name suggests, the generic network restoration problem seeks for the best recovery strategy for a given perturbed network. As a case study, a disrupted network from a pot-disaster environment is tackled. Proposed frameworks are demonstrated on the real world disrupted road networks of different cities in USA having various topological properties.

380