Informs Annual Meeting 2017

TA05

INFORMS Houston – 2017

3 - Choice of E-waste Recycling Standard under Recovery Channel Competition Gokce Esenduran, Ohio State University, 2100 Neil Avenue, 656 Fisher Hall, Columbus, OH, 43210, United States, esenduran@gmail.com, Yen-Ting (Daniel) Lin, Wenli Xiao, Minyue Jin There are two recycling standards for e-waste recyclers that are different in stringency: E-stewards and R2. E-stewards is more stringent and costlier but attracts more e-waste from environmentally conscious donors. We model the competition between recovery channels and examine how secondary market sales and economies of scale affect recyclers’ choice. We find that a recycler may choose e-Stewards only when facing competition. Our results also show that recyclers with strong scale economies in e-waste processing choose e-Stewards only when incurring processing costs significantly higher than R2 would require. 4 - To Recycle or Not: An Analysis of the Environmental and Financial Impact of Recycling Hailong Cui, University of Southern California, Marshall School of Business, Data Sciences and Operations, Los Angeles, CA, 90089- 0809, United States, Hailong.Cui.2019@marshall.usc.edu, Greys Sosic We evaluate the conditions under which recycling of simple products (made of common materials) has positive impact across three dimensions: GHG emissions, operational costs, and societal costs. Our study contains both an analytical model and empirical validation of our results. We then consider different decision makers when decision variables are collection and yield rates: we start with centralized models wherein a single entity (local government, manufacturer, or independent recycler) determines both values, and then we move to decentralized models in which collection and yield rates are chosen by two different parties. We also discuss incentive mechanisms for achieving social optima. 320B Behavioral, Incentive and Policy Issues in Healthcare Sponsored: Health Applications Sponsored Session Chair: Tinglong Dai, Johns Hopkins University, Baltimore, MD, 21202, United States, dai@jhu.edu Co-Chair: Song-Hee Kim, University of Southern California, Los Angeles, CA, 90089, United States, songheek@marshall.usc.edu 1 - Search Among Queues under Quality Differentiation Luyi Yang, Johns Hopkins University, Baltimore, MD, United States, luyi.yang@jhu.edu, Laurens G.Debo, Varun Gupta We examine a setting in which customers looking for service providers face search frictions and service providers vary in quality and availability. We find that reducing either the search cost or customer arrival rate may strictly increase the average waiting time in the system as customers substitute toward high-quality service providers. We discuss policy implications of our results in the context of public surgical waits. 2 - Upcoding in Medicare Reimbursement for Hospital-acquired Infections Joel Goh, Harvard Business School, Morgan Hall 420C, Soldiers Field Road, Boston, MA, 02163, United States, jgoh@hbs.edu, Hamsa Sridhar Bastani, Mohsen Bayati Since 2008, Medicare has levied financial penalties to providers for hospital- acquired infections (HAIs). However, Medicare relies on providers’ self reported HAIs to assess these penalties. Consequently, providers may upcode, i.e., mis- report HAIs in a manner that increases reimbursement. One form of upcoding involves incorrectly classifying HAI conditions as being present-on-admission (POA). In this empirical study, we use nationally-representative claims data to investigate the whether such POA upcoding occurs. Our results suggest that the level of state-level regulation and enforcement is an important determinant of POA upcoding. 3 - Designing Response Supply Chain Against Bioattacks Peter Yun Zhang, Massachusetts Institute of Technology, Cambridge, MA, United States, pyzhang@mit.edu, David Simchi-Levi, Nikolaos Trichakis We study the problem of designing supply chain against bioattacks. We propose a model that integrates key facets of the problem in a way not done before, with decision variables spanning medical countermeasure stock level, capacity, shipment, and dispensing. We also explicitly capture the interaction between defender and attacker. We present a tractable and provably strong heuristic capable of solving large networks with high fidelity (100,000s nodes). Compiling data on supply, population, lead times, health effects, and costs, we analyze the design of a comprehensive biodefense supply chain in the United States and generate new policy insights. TA05

4 - A Contract Mechanism for the Opioid Prescription Epidemic Alireza Boloori, Arizona State University, 505 W. Baseline Rd, Apt 2158, Tempe, AZ, 85283, United States, alireza.boloori@asu.edu, Soroush Saghafian, Stephen J. Traub Hospital EDs are in part responsible for the overprescription of opioid painkillers. To analyze the impact of different contracts (between an insurer (e.g., Medicare and a hospital) on the opioid prescription in EDs, we propose a principal-agent model where pharmaceutical incentives from opioid manufacturers cause asymmetric information (i.e., adverse selection) in the model. Utilizing data from Mayo Clinic, we establish a data-driven contracting approach to not only alleviate the negative consequences of opioids, but also address patients who really need these medications. We also provide implications for healthcare policy makers. 320C Healthcare Decision Making under Uncertainty Sponsored: Health Applications Sponsored Session Chair: Hadi El-Amine, George Mason University, George Mason University, Fairfax, VA, 22030, United States, helamine@gmu.edu 1 - Adaptive Risk-based Pooling in Public Health Screening Hrayer Y. Aprahamian, Virginia Tech, 1145 Perry Street, Blacksburg, VA, 24060, United States, ahrayer@vt.edu, Ebru Korular Bish, Douglas R.Bish Pooled testing is commonly used in public health screening for classifying subjects in a large population as positive or negative for an infectious or genetic disease. Pooling is especially useful when screening for low prevalence diseases under limited resources. We propose and study an adaptive risk-based pooling scheme, which considers important test and population level characteristics often over- looked in the literature. We characterize important structural properties of optimal subject assignment policies, and provide key insights. Our case study demonstrates the effectiveness of the proposed pooling scheme, with false classifications reduced substantially over current policies. 2 - Multi-objective Two-stage Stochastic Programming for Adaptive Interdisciplinary Pain Management with PLN Transition Models Jay Michael Rosenberger, University of Texas-Arlington, Arlington, TX, 76010, United States, jrosenbe@uta.edu, Gazi Md Daud Iqbal, Victoria Chen, Robert Gatchel This research uses a two-stage stochastic programming approach to optimize personal adaptive treatment strategies for pain management. The goal is to generate adaptive treatment strategies using statistics based optimization approaches that can be used by physicians to prescribe treatment to the patients. Five pain outcomes has been considered in objective function. This research uses Piecewise Linear Networks (PLN) to represent transition models. A mixed integer linear program is developed to integrate those PLN transition models into an optimization problem. 3 - Comparative Effectiveness of Incorporating a Hypothetical DCIS Prognostic Marker into Breast Cancer Screening Mehmet Ergun, University of Wisconsin, 7429 Old Sauk Road, Madison, WI, 53717, United States, mergun@wisc.edu, Oguzhan Alagoz, Amy Trentham-Dietz, Ronald E. Gangnon, Natasha Stout, John M.Hampton, Kim Dittus, Ted A. James, Pamela M. Vacek, Sally Herschorn, Elizabeth S. Burnside, Anna N. Tosteson, Donald L. Weaver, Brian L. Sprague Due to the lack of a reliable test to identify a non-progressive disease, most of the ductal carcinoma in situ (DCIS) cases are treated with similar treatment regimens to localized invasive breast cancers. Thus, most of the patients with a non- progressive disease receive unnecessary treatments that result in decrements in their quality of life as well as an extra financial burden. To assess the potential impact of a hypothetical test that identifies non-progressive DCIS cases, we used the University of Wisconsin Breast Cancer Simulation model. Our results show that a perfect test would reduce the treatment costs as much as 40% and lead to a modest increase in discounted quality-adjusted life years. 4 - Robust Post-donation Blood Screening under Prevalence Rate Uncertainty Hadi El-Amine, George Mason University, 4310 Cotswolds hill ln, Fairfax, VA, 22030, United States, helamine@gmu.edu, Ebru Korular Bish, Douglas R. Bish Blood product safety, in terms of being free of transfusion-transmittable infections, is crucial. Under prevalence rate uncertainty, various objective functions, including minimization of a mean-variance objective and minimization of the maximum regret, were considered in order to determine a “robust” post- donation blood screening strategy that minimizes the risk of releasing an infected unit of blood into the blood supply. Efficient and exact algorithms are provided. TA06

270