2016 INFORMS Annual Meeting Program

SD24

INFORMS Nashville – 2016

SD22 107B-MCC Disaster Relief Supply Chains and Operations Sponsored: Public Sector OR Sponsored Session Chair: Felipe Aros-Vera, Ohio University, Stocker Center 277, 1 Ohio University, Athens OH, OH, 45701, United States, aros@ohio.edu 1 - Pre-positioning Emergency Relief Items Before A Typhoon With An Uncertain Trajectory Joline Uichanco, University of Michigan, Ross School of Business, jolineu@umich.edu We describe a collaborative work with the Philippine government on a pre- positioning model in preparation for an oncoming typhoon. Pre-positioning relief aid before a typhoon is challenging due to the uncertainties in locations and quantities of future demand. We develop a prediction model for the number of affected population by fitting a dataset of typhoon effects to a hierarchical linear model. Our model reveals a significant relationship between wind speed and affected population. We propose a bi-objective stochastic pre-positioning model which balances fairness and effectiveness of the pre-positioning strategy. 2 - Ecuador Earthquake Relief Support: Observations From Fieldwork Research Johanna Amaya, Rensselaer Polytechnic Institute, Troy, NY, United States, amayaj@iastate.edu, Johanna Amaya, Iowa State University, Ames, IA, 50011, United States, amayaj@iastate.edu, Cinthia Perez Siguenza, Jose Holguin-Veras This talk presents an overview of the disaster response logistics that took place after the earthquake in Ecuador. The talk discusses the preliminary results of the fieldwork research conducted by the authors in the aftermath of the disaster. 3 - Objectives’ Misalignment In Humanitarian Operations: The Role Of Earmarking Effectiveness of humanitarian programs depends both on the donors’ willingness to support the program and on the program implementation by the international humanitarian organization (IHO). Donors donate with the aim of reaching as more beneficiaries as possible. IHOs also have the same objective, but face constraints on how they can use the available funds. A big constraint comes from the donors themselves, who often earmark their funding. In this paper, we analyze the donors’ and IHOs’ decision-making in an effort to shed more light on how decisions for earmarking are taken. The aim is to give recommendations to the IHOs on how to align donors’ objectives to theirs. 4 - Willingness-to-pay Models On Post-disaster Environments Diana Ramirez-Rios, Research Assistant, Rensselaer Polytechnic Institute, Troy, NY, 12180, United States, ramird2@rpi.edu Jose Holguin-Veras, Johanna Amaya, Trilce Marie Encarnacion, Shaligram Pokharel, Victor Cantillo, Luk Wassenhove This paper introduces an economic valuation for the level of anxiety of an individual under deprivation conditions as anxiety is well-known measure of psychological distress in a community. More specifically, this research estimated the willingness-to-pay for water of individuals who have been affected by disasters, under different scenarios of deprivation and expectation. The level of anxiety is measured by the effect that the expected time to normality introduces to WTP, and results indicate that that as the time to recover increases, the level of suffering increases. “ SD23 108-MCC Applications in Physician Scheduling Sponsored: Health Applications Sponsored Session Chair: Andreas Fügener, Universität zu Köln, 123334, Germany, andreas.fuegener@uni-koeln.de 1 - Decision Support For Physician Rostering: Development Of Models And Implementation Of Software Jens Brunner, University of Augsburg, jens.brunner@unikat.uni-augsburg.de, Andreas Fuegener In order to cope with steadily increasing healthcare costs, hospitals try to schedule their physicians efficiently and effectively. We consider a scheduling problem at large teaching hospitals in Germany. We formulate mixed-integer linear programming models for duty- and workstation assignments subject to union contracts as well as individual agreements of the physicians. To promote for job satisfaction we take into account fairness and preferences. We present the status of the software development and discuss lessons learned from the project and highlight some barriers when it comes to implementation of decision support systems in practice. Laura Turrini, Kuehne Logistics University, laura.turrini@the-klu.org, Maria Besiou

2 - Neonatal Physician Scheduling At The University Of Tennessee Medical Center Charles E Noon, University of Tennessee-Knoxville, Knoxville, TN, United States, cnoon@utk.edu, Melissa R Bowers, Wei Wu, Kirk Bass The default approach for scheduling hospital coverage is to distribute the various types of shifts equally among the covering physicians. This “equality” approach insures that each physician works his/her fair share of overnights, weekends, etc. We present a new model that incorporates individual shift-type preference so that each physician attains a schedule that is equivalent or superior to his/her “equality” schedule. We formulate and solve the model as a mixed integer program. We demonstrate its benefits by using the approach to schedule hospital coverage for a neonatology group. 3 - Equitable Scheduling Of Resident Shifts Hernan Abeledo, George Washington University, abeledo@gwu.edu, Anthony Coudert Creating shift schedules for resident physicians is a notoriously difficult task that is typically done manually by the chief residents. Shift assignments need to observe a large number of rules while populating a complex schedule structure. A key goal is that the schedule be perceived as fair by all residents. We present an integer programming model used to schedule anesthesiology residents at the George Washington University Hospital. The fairness objective is addressed through a point system proposed by the residents. 4 - Re-scheduling Of Physicians In Case Of Unexpected Absences Andreas Fügener, University of Cologne, andreas.fuegener@uni-koeln.de, Christopher Gross, Jens Brunner Scheduling physicians is a complex task as legal requirements, levels of qualification, and preferences for different working hours should be considered. Unplanned absences, e.g. due to illness, additionally drive the complexity. In this study, we discuss an approach to deal with the following trade-off: Changes to the existing plan should be kept as small as possible. However, an updated plan should still meet the requirements regarding work regulation, qualifications needed, and employee preferences. We present a mixed-integer programming model to create updated plans following absences of scheduled personnel and apply it to real-life data from a German university hospital. 109-MCC New Directions in Non-Market Strategies Invited: Strategy Science Invited Session Chair: Jason Snyder, University of Utah, Eccles School, Salt Lake City, UT, 9, United States, Jason.snyder@eccles.utah.edu 1 - Locked In? Noncompete Enforceability And The Mobility And Earnings Of High-tech Workers Jin Woo Chang`, University of Michigan, Ann Arbor, MI, United States, jinwooch@umich.edu, Natarajan Balasubramanian, Mariko Sakakibara, Jagadeesh Sivadasan, Evan Starr We use matched employer-employee data from 30 U.S. states to examine how the enforceability of noncompete contracts affects the length of job spells and the level of wages. Exploiting inter-state variation in the degree of enforceability and controlling for worker-, job-, and state-level characteristics, we find that a unit standard deviation increase in enforceability is associated with a 3.6% increase in the length of job-spells for high-wage workers in technology industries. We also find persistent wage suppressing effects that last throughout their employment history. Together, these are consistent with noncompetes reducing the bargaining power of employees relative to their employers. 2 - On A Firm’s Optimal Response To Pressure For Gender Pay Equity David Ross, University of Florida, 55, Gainesville, FL, 32611, United States, David.Ross@warrington.ufl.edu David Anderson, Cristian Dezsö, Margret Bjarnadottir We present a theory of how a firm would respond to pressure for gender pay equity by strategically distributing raises and adjusting its organizational structure. Using mathematical reasoning, simulations, and data from a real employer, we show that (a) employees in low-paying jobs and whose job-related traits typify men at the firm are most likely to get raises; (b) counterintuitively, some men will get raises and giving raises to certain women would increase the pay gap; (c) a firm can reduce the gender pay gap as measured by a much larger percentage than the overall increase in pay to women at the firm; and (d) “ghettoizing” women in select jobs can help a firm reduce its pay gap. SD24

101