INFORMS 2021 Program Book

INFORMS Anaheim 2021

ME10

3 - Optimal Character Selection in DND Michael A. Perry, Fresno State University, Fresno, CA, United States, Aaron Bradley Hoskins The research uses a Monte Carlo simulation to determine character survival rate in a typical one day of adventuring in Dungeons and Dragons. The Duelist Algorithm is used as an outer loop to optimize the survival rate of the adventuring party. Comparisons to other metaheuristics are also provided. 4 - Soccer Penalty Shootouts: Network Analysis and Performance Mechanisms Nils Rudi, Yale School of Management, New Haven, CT, 06511, United States Modeling soccer penalty shootouts as a probability network facilitates analysis of several questions, including the impact of alternative sequences such as ABBA. I prove several properties of this model that justify the underlying markov assumption. Using a large dataset of penalty shootouts, I study potential performance drivers and mechanisms combining the markov model and traditional econometrics. ME10 CC Room 304B In Person: Management of Service Systems General Session Chair: Wei Liu, University of North Carolina at Chapel Hill, United States 1 - Optimal Scheduling of Proactive Care with Patient Deterioration Jing Dong, Columbia University, New York, NY, 10027-6945, United States, Yue Hu, Carri Chan Healthcare systems are limited resource environment where scarce capacity is often reserved for the most critical patients. However, there has been a growing interest in the use of proactive care when a less urgent patient may become urgent while waiting. On one hand, providing service for patients when they are less urgent could mean that fewer resources are needed to serve them. On the other hand, utilizing limited capacity for patients who may never need the level of care in the future takes the resource away from the more urgent patients. To understand this tension, we propose a queueing model with two patient classes: moderate and urgent, and allow patients to change classes while waiting. We characterize how moderate and urgent patients should be prioritized for service when proactive care for moderate patients is an option. 1 - Joint Staffing and Admission Control Problem under Different Levels of Information Wei Liu, University of North Carolina at Chapel Hill, Chapel Hill, NC, United States, Vidyadhar Kulkarni We consider a joint staffing and admission control problem in a multi-server queueing system under three different levels of information, namely minimal, partial and full information about the state of the queueing system. Our major contribution lies in the combination of the admission control, staffing problem, and information levels. The system earns a unit reward from serving a customer if her queueing time is no more than a fixed threshold and each server costs a fixed amount per unit time. Under each information case, we derive the optimal admission policy and optimal staffing level by maximizing the profit. We also study the criterion of maximizing the revenue rate per server and show the connection between these two criteria in determining the optimal staffing level. Finally, we compare the system performance under different levels of information.

ME07 CC Room 303B In Person: Emerging Topics in Supply Chain General Session Chair: Woonam Hwang, University of Utah, Salt Lake City, UT, 84112- 8939, United States Co-Chair: Anyan Qi, The University of Texas at Dallas, Richardson, TX, 75080-3021, United States Co-Chair: Andre Du Pin Calmon, Scheller College of Business, Georgia Institute of Technology, Atlanta, GA, 30308, United States 1 - Strategic Overcapacity in Live-streaming Platformselling Anyan Qi, The University of Texas at Dallas, Richardson, TX, 75080-3021, United States, Suresh P. Sethi, Liqun Wei, Jianxiong Zhang We study the capacity investment strategy of a manufacturer who sells his product on a live-streaming shopping platform. The manufacturer first decides the production capacity, then the platform decides her commission, and finally the manufacturer sets the retail price. The platform has an informational advantage about the product demand due to proximity to the market and accessibility to the sales data of similar products. The manufacturer without a direct access to the demand information tries to infer it from the commission decision of the platform, which results in a signaling game. Interestingly, the manufacturer may strategically install a strictly higher capacity than any demand to be realized. The overcapacity also benefits the manufacturer by driving down the commission charged by the platform when observing a small market potential due to the signaling effect. 2 - Competition and Innovation Zhibin (Ben) Yang, Universtiy of Oregon, Eugene, OR, 97403- 1205, United States, Jie Ning We study strategic interaction between an innovation-leader firm and a follower firm that both sell to and compete in the same market. The follower firm is less innovative and has the option to source from the leader firm. We analyze a multi- stage game and characterize the Sub-game Perfect Nash Equilibrium. 3 - Ambulance Platforms to Improve Response Times for Emergency Calls In Developing Countries Many developing countries lack the infrastructure for emergency response of the developed world. Often, the problem is not the lack of capacity, but the lack of coordination. To solve this coordination problem, multiple companies, for example Flare in Kenia and StanPlus in India, have started ambulance platforms to bring together demand and supply for emergency care. We study how the operational process of these platform services differs from traditional EMS providers and help to improve the service level of these companies. ME08 CC Room 303C In Person: SpORts IV General Session Chair: Nils Rudi, Yale School of Management, New Haven, CT, 06511, United States 1 - Determining Expected Win Percentage in the Indian Premier League Aaron Bradley Hoskins, California State University, Fresno, CA, 93710, United States The Indian Premier League (IPL) is the most lucrative professional cricket league in the world. This research investigates determining the expected winning percentage of a team in the league based on underlying team performance measures. 2 - Predicting Postseason Success in Major League Baseball using Data Envelopment Analysis Christopher Gaffney, Associate Clinical Professor, Drexel University, Philadelphia, PA, United States Translating in-season performance to post-season success is a well-known issue in professional sports. While there are several well-known methods to predict regular season performance in Major League Baseball, postseason prediction is made more difficult by the increased variability that comes with a compressed schedule. In this paper we use data envelopment analysis to study the impact of efficient resource utilization on postseason performance. Pieter van den Berg, Rotterdam School of Management, Erasmus University, Rotterdam, 3062 PA, Netherlands

89