INFORMS 2021 Program Book

INFORMS Anaheim 2021

SD23

2 - Negative Spillover on Service Level Across Priority Classes: Evidence From a Radiology Workflow Platform Bernardo (Bernie) F. Quiroga, Pontificia Universidad Catolica de Chile, School of Management, Macul, Santiago RM, Chile, Timothy Chan, Nicholas Howard, Saman Lagzi, Gonzalo Romero We study a radiology workflow platform connecting off-site radiologists with hospitals: Tasks are selected from a common pool, and the service level is characterized by meeting priority-specific turnaround time targets. Imbalances between pay and workload of different tasks could result in higher priority tasks with low pay relative to workload receiving poorer service than low priority tasks. We analyze, using IV regressions, if low priority tasks with a high pay-to- workload ratio have shorter turnaround times, and if low priority tasks with high pay-to-workload increase turnaround times and probability of delay of higher priority tasks (negative spillover). We find evidence that turnaround time decreases in pay-to-workload for lower priority tasks, increases in workload for high priority tasks, and also presence of a negative spillover effect. 3 - The Success of Women Leadership in Fighting Covid-19: Evidence from U.S. Nursing Homes Poyraz Bozkurt, Purdue University, West Lafayette, IN, United States, Susan F. Lu, Lauren Xiaoyuan Lu Using the U.S. nursing home data, we investigate the impact of women leadership on Covid-19 infection and death rates in nursing homes. We observe that a higher number of women directors in management teams results in fewer Covid- 19 infections. We further investigate the mechanisms by analyzing staff planning, PPE shortages and weekly visitors. Our findings suggest that women leadership leads to higher staffing and quality ratings. Moreover, nursing homes led by female directors are more likely to have weekly availability of PPE supply. While these results are significant for all nursing homes, our analysis reveals that the impact of female managers is weaker in for-profit nursing homes. 4 - Learning in Drug Shortages Hyun Seok (Huck) Lee, Korea University Business School, Seoul, 97333-3235, Korea, Republic of, Junghee Lee, In Joon Noh In this study, we investigate whether pharmaceutical manufacturing plants learn from their own drug shortage instances. Specifically, we examine if more drug shortages recovery at a plant lead to quicker recovery from its subsequent shortages. We also investigate factors that might affect this learning. Our findings will have policy implications for the FDA and will also contribute to the academic literature on learning. SD22 CC Room 204B In Person: OR Methods for Health Policy Design General Session Chair: Saumya Sinha, Rice University, Houston, TX, 77005-1827, United States 1 - Multi-year Optimization of Malaria Intervention: A Mathematical Model Susan E. Martonosi, Professor, Harvey Mudd College, Claremont, CA, 91711-5901, United States, Harry Dudley, Abhishek Goenka, Cesar Orellana Malaria is a mosquito-borne, lethal disease that affects millions and kills hundreds of thousands of people each year, mostly children. In this paper, we couple a susceptible-infected-recovered compartment model for the disease dynamics with an integer linear program to allocate malaria interventions across geographic regions and time, subject to budget constraints, with the aim of minimizing the number of person-days of malaria infection. The model provides a qualitative decision-making tool to weigh alternatives and guide malaria eradication efforts. A one-size-fits-all campaign is found not to be cost-effective; it is better to consider geographic variations and changes in malaria transmission over time when determining intervention strategies. 2 - Combination Chemotherapy Optimization Temitayo Ajayi, Nature Source Improved Plants, Ithaca, NY, 77004, United States, Dave Fuller, Andrew J. Schaefer, Mohammad Hosseinian Chemotherapy is one of the primary modalities of cancer treatment. Chemotherapy drug administration is a complex problem that often requires expensive clinical trials to evaluate potential regimens. One way to better inform future trials is to build reliable models that illustrate how a patient may react to specified drugs and doses. Previous chemotherapy optimization models have relied on optimal control, which does not lend itself to discrete considerations such as doses via pills and rest periods. In this paper, we develop mixed-integer linear programming models for combination chemotherapy that incorporate various important operational constraints. We also address uncertainty in the tumor heterogeneity with a chance constraint.

3 - Incentives in Outcome-based Regulation for Organ Transplantation Saumya Sinha, Rice University, Houston, TX, 77005-1827, United States, David Mildebrath, Taewoo Lee, Andrew J. Schaefer Federal agencies use outcome-based regulatory criteria for oversight of transplant programs, aiming to incentivize programs to improve their post-transplant outcomes. However, clinical evidence indicates that the regulations may induce programs to reject medically suitable patients to avoid penalization. We present a game-theoretic model of transplant programs to analyze the incentives created by these regulations. We demonstrate that excessively harsh penalization, more so than other factors, incentivizes programs to engage in adverse patient selection. We propose an alternative pay-for-performance reimbursement scheme which penalizes underperforming programs and pays a bonus to programs with above- average outcomes. The proposed scheme can incentivize programs to improve post-transplant outcomes without inducing adverse patient selection. SD23 CC Room 204C In Person: Service Workforces General Session Chair: Vincent Slaugh, Cornell University, Genoa, NY, 13071, United States 1 - Staffing for Housekeeping Operations Buyun Li, Indiana University, Bloomington, IN, United States, Vincent Slaugh We develop an analytical model of hotel housekeeping to minimize staffing costs and guest wait times for room readiness. We show structural properties, including discrete convexity, of the single-day rostering problem for room attendant shift start times. Using data from a hotel, heuristics enabled by these save up to 17% of total costs compared to a common industry staffing heuristic. We also describe strategies for hotels facing a staffing shortage, including the use of part-time workers. 2 - Optimal Return Time Window with Consumer Learning Punya Chatterjee, Pennsylvania State University, State College, PA, United States, Aydin Alptekinoglu, Nicholas C. Petruzzi In this paper, we analytically study a retailer’s decision of the length of return time window when consumers update their product valuation over time as they consume a product. We develop a model consisting of a profit-maximizing retailer who sets the length of its return time window, a product which has a finite life- time, and a forward looking consumer who needs to consume the product to understand how much they value the product (e.g., electronics). Our results can guide retailers to select return time windows for various product categories and different consumer types in the case that the consumers learn over time. 3 - On Designing a Socially-optimal Expedited Service and its Impact on Individual Welfare Our research is motivated by the expedited security check at US airports (TSA PreCheck). We consider the problem faced by a welfare-maximizing service provider who must make a decision on how to split a fixed capacity between a standard service and an expedited service. The service is mandatory. Choosing the expedited variant requires enrollment at a fixed cost per period. Customers are strategic and have the same cost of waiting, but are heterogeneous in the rate at which they use the service. We show customers’ strategic behavior in equilibrium is uniquely determined by the provider’s allocation decision. We use this result to solve for the socially optimal allocation. We show that even when customers behave strategically, an expedited service offered in parallel to a standard service can not only increase overall welfare, but also do so for each customer individually. 4 - Congestion, Conflict, and Coordination: Contracting with a Food Delivery Platform Andrew E. Frazelle, The University of Texas at Dallas, Dallas, TX, 75205-3685, United States, Pnina Feldman, Robert Swinney In a stylized model of a restaurant as a congested service system, we explore various contractual arrangements between the parties. We find that the commonly-used, traditional revenue sharing contract, in which the platform takes a percentage cut of each delivery order, fails to coordinate the system because the platform does not internalize the effect of its pricing on dine-in revenues. By contrast, a no-contract arrangement, in which the platform pays menu price on each unit, protects the restaurant’s revenue from being cannibalized by lower-margin delivery orders. Unfortunately, it too leaves potential revenue on the table. We propose an alternative, practical coordinating contract. As well as coordinating the system, our contract protects restaurant margins by ensuring that the restaurant receives no less per order on delivery than dine-in. Ricky Roet-Green, Simon Business School, University of Rochester, Rochester, NY, 14534-2883, United States, Aditya Shetty

37