Informs Annual Meeting Phoenix 2018

INFORMS Phoenix – 2018

WA57

2 - Knowledge Outsourcing for Competing Buyers Jaeseok Lee, Georgia Institute of Technology, 800 West Peachtree NW, Atlanta, GA, 30308, United States, Cheryl Gaimon, Karthik Ramachandran We study the knowledge development and outsourcing decisions of competing firms (buyers), who can obtain knowledge from a common supplier. We identify a condition whereby buyers benefit from knowledge outsourcing. We also discuss how the supplier is better off if he serves only one buyer. 3 - Market Entry and Location Choice under Market Uncertainty Wenxin Xu, The Hong Kong Polytechnic University, Hong Kong When a new bicycle-sharing company considers pursuing the growth opportunities in emerging markets, e.g, the second or third-tier cities, it often grapples with those questions such as: When should I enter this new market? Where should I locate the docks to serve the customers while avoiding the fierce competitions? This paper provides some managerial insights into the entry timing and location choice under market uncertainty for those companies with a dynamic game-theoretical model. 4 - Initial Stage Success Strategies for Mobile Applications Moonwon Chung, University of South Carolina, Cayce, SC, 29033, United States, Luv Sharma, Manoj Malhotra This study examines the implications of operations decisions in the hypercompetitive context of software development, dissimilation, and sustained service delivery. Specifically, we estimate performance impact of user engagement impacted by the choice of mobile app feature decisions, product launch timing decisions, decentralized value chains, and organizational focus. A panel data of Top 500 ranked mobile game apps in the US was extracted from app market APIs in daily observations over 3 years. We perform econometric analyses to reveal software development and value chain configuration implications for managers. n WA56 West Bldg 101A Optimization in Cancer Treatment Sponsored: Health Applications Sponsored Session Chair: Ali Ajdari, Massachusetts General Hospital & Harvard Medical School, Seattle, WA, 98199, United States 1 - A Progressive Hedging Approach for Chemotherapy Appointment Scheduling Nur Banu Demir, Middle East Technical University, Ankara, Turkey, Serhat Gul, Melih Celik Chemotherapy appointment scheduling in oncology clinics is a challenging combinatorial optimization problem due to the assignment of patients to nurses and chairs. Furthermore, the patient appointment times of a daily patient list must be determined under the uncertainty of the infusion preparation and treatment times. The problem is formulated as a two-stage stochastic mixed integer programming model that minimizes the expected weighted sum of nurse overtime and patient waiting time. A Progressive Hedging Algorithm is applied in order to obtain appointment times for patients. Computational experiments are performed using real data of a major university hospital. 2 - Including Edge-penalization Aperture Control Methodologies into VMAT Optimization Wilmer Henao, University of Michigan, Ann Arbor, MI, United States, Marina A. Epelman, Martha Matuszak, Kelly Younge, Edwin Romeijn Complex beam aperture shapes consisting of excessive perimeter per unit of area can potentially give rise to dose inaccuracies in VMAT radiation therapy treatments. Standard practice does not explicitly take aperture shape into account at the planning optimization stage. We intend to correct this issue by approximating previously well-established edge penalties, and explicitly incorporating them into the optimization using a column-generation based heuristic. Our algorithm was tested in real cases obtaining a reduction of aperture complexity, an ensuant increase in dose accuracy and a negligible decrease in treatment quality as measured by Dose-Volume Histograms. 3 - A Novel Optimal Stopping Approach to Formulate the Radiation Therapy Problem Ali Ajdari, Research fellow, Massachusetts General Hospital & Harvard Medical School, 125 Nashua st, Boston, MA, 02114, United States, Thomas Bortfeld In the fractionated radiation therapy problem, the treatment is delivered over the course of several sessions. The length of the treatment is usually determined by one-size-fits-all guidelines and is therefore fixed prior to the beginning of radiotherapy course. In some cases however it might be optimal to stop the treatment early to avoid severe normal tissue toxicities and/or over-treating the patient. Using tools in dynamic programming and control theory, we formulate the problem as an optimal stopping model and try to find the optimal time to stop the treatment, based on the observed tumor and normal tissue response.

n WA57 West Bldg 101B Resilient Pharmaceutical Supply Chains Sponsored: Health Applications Sponsored Session Chair: Rozhin Doroudi, Northeastern University, Boston, MA, United States 1 - Policies to Improve Pharmaceutical Supply Chain Resiliency: A Multi-stage Stochastic Programming Approach Emily L. Tucker, University of Michigan, Ann Arbor, MI, 48105, United States, Mark S. Daskin, Wallace J. Hopp, Burgunda V. Sweet Drug shortages continue to be a major, though little-discussed, public health crisis. They are often caused by disruptions to non-resilient pharmaceutical supply chains. We present a new multi-stage stochastic programming model of the supply chain design problem when components may be disrupted. We discuss the application of a variant of the Stochastic Dual Dynamic Programming algorithm and analyze how policy proposals would affect the risk of shortages under a variety of market conditions. 2 - Minimization of Drug Shortages in the Pharmaceutical Supply Chains: A Simulation-Based Analysis of Drug Recall Patterns and Inventory Policies Rana Azghandi, Northeastern University, Boston, MA, 02215, United States, Jacqueline Griffin, Mohammad S. Jalali The drug shortage crisis in the last decade not only increased health care costs but also put patients’ health at risk in the United States. Managing pharmaceutical supply chains is very complex, as inevitable disruptions occur to these supply chains (exogenous factors), which are followed by decisions that members make after such disruptions (internal factors). The current research employs a system dynamics simulation model to examine the effects of different disruption profiles, e.g., the propagation in time and space, and the interactions of decision makers on drug shortages to ascertain how these shortages can be mitigated by changing the inventory policy decisions. 3 - Building Up Resilience in a Pharmaceutical Supply Chain through Inventory, Dual Sourcing, and Agility Capacity Florian Lucker, Cass Business School, 106 Bunhill Row, London, EC1Y 8TZ, United Kingdom, Ralf W. Seifert In this talk we investigate the relationship between the three operational risk mitigation measures inventory, dual sourcing, and agility capacity by modelling a drug manufacturing firm. We quantify the decrease in inventory levels in the presence of dual sourcing and agility capacity. Furthermore, we show how to determine optimal inventory and dual sourcing decisions holistically. Within our modeling framework we introduce an operational metric that quantifies supply chain resilience. 4 - Drivers of Drug Shortages: Theoretical and Empirical Investigations Iva Petrova Rashkova, Washington University-St Louis, Campus Box 1156, Business School Faculty, St Louis, MO, 63130, United States, Panos Kouvelis We use the drug shortages FDA data during 2012-2016 to identify the driving forces of drug shortages. We first develop a stylized model illustrating the relationship between drug shortages and changes in the product portfolio of a drug manufacturer. Our key insight is that the potential new drugs, regardless of whether they end up being successful or not, drive drug shortages. We then use data on new brand and generic drugs to show their usefulness in predicting manufacturer’s drug shortages. We compare linear regression and random forest predictive models and recommend the latter based on its lower MAPEs both in- samlpe and out-of-sample. 5 - Identifying Critical Components of a Multi Agent Drug Supply Chain Rozhin Doroudi, Northeastern University, Boston, MA, 02116, United States, Ozlem Ergun The growing epidemic of drug shortages in the US indicates the fragility of drug supply chain in the face of disruptions. Identifying most critical components in the supply chain can assist policy makers in mitigating these shortages. In this study we propose a node criticality measure for underlying network of a drug supply chain in which nodes are agents and arcs are material flow between them. Unlike most node criticality measures in network science literature this measure does not abstract away domain-specific functions of network components. Therefore, it can be used in predicting the effect that disruption of one node can have on the overall supply chain performance.

431