2015 Informs Annual Meeting

WA63

INFORMS Philadelphia – 2015

WA63 63-Room 112B, CC Operations Management III Contributed Session Chair: Sang Jo Kim, MIT-Zaragoza International Logistics Program, Calle Bari 55, Edificio Nayade 5, Zaragoza, 50197, Spain, sjkim@zlc.edu.es 1 - A New Approach to the Multi-Product Newsvendor Problem with Customer Choice Amr Farahat, Washington University in St. Louis, One Brookings Drive, St. Louis, MO, 63104, United States of America, farahat@wustl.edu, Joonkyum Lee We present a methodology that yields upper bounds and feasible solutions to the multiproduct newsvendor problem with customer choice. The methodology is based on an approximate Jordan decomposition of the state transition matrix. Specializations include a decomposition by customer into a sequence of assortment optimization problems and a decomposition by product into a collection of independent newsvendor problems. 2 - Stackelberg-game Analysis of Retailers’ Joint Procurement Tianyuan Wang, Lingnan (University) College of Sun Yat-Sen University, NO.135, Xingang West Road, Guangzhou, China, wty-2@163.com, Ke Fu We consider two retailers who face a Stackelberg game in joint procurement. They have to determine the optimal inventory and pricing strategies in the presence of both cooperation and competition. We show that both retailers prefer joint purchasing if they have similar market penetrations. We demonstrate how different problem parameters affect the game outcomes. 3 - Consumer Showrooming and the Length of Product Line Yilong Luo, PhD Candidate, Illinois Institute of Technology, 6716 Idaho Ave., Hammond, IN, 46323, United States of America, luoyilong@gmail.com, Jiong Sun It is known that consumers’ showrooming behavior, i.e., inspecting products at brick-and-mortar stores and then making a purchase at a competing online store, can result in the brick-and-mortar store losing potential customers. We show that such showrooming behavior can soften competition and hence can benefit both retailers. We also show that this showrooming behavior may or may not induce the brick-and-mortar retailer to reduce the length of the product line it carries. 4 - The Optimal Dynamic Policy for Periodic Stocking and Bundling Sang Jo Kim, MIT-Zaragoza International Logistics Program, Calle Bari 55, Edificio Nayade 5, Zaragoza, 50197, Spain, sjkim@zlc.edu.es, Youyi Feng, Jianjun Xu We consider a firm that offers two individual products and a bundle of them. They periodically order the two products as well as bundle them aiming to maximize the total discounted profit over a finite planning horizon. We assume that the customer demand is random, unmet demand is backlogged and unsold items are carried over to the next period. Optimal joint decisions on ordering and bundling follow a well-structured policy that also draws significant managerial insights. WA64 64-Room 113A, CC Risk-Based Decision Making in Interdependent Systems Sponsor: Decision Analysis Sponsored Session Chair: Hiba Baroud, Vanderbilt University, 400 24th Avenue South, Nashville, TN, 37205, United States of America, hiba.baroud@vanderbilt.edu 1 - Multi-criteria Inoperability Analysis of Commodity-specific Dock Disruptions of Inland Waterway Port Mackenzie Whitman, Graduate Student, University of Oklahoma, 1018 Timber Ridge Road, Harrah, OK, 73045, United States of America, mgwhitman@ou.edu, Hiba Baroud, Kash Barker Decision making for managing risk to infrastructure systems requires accounting for impacts of disruptions that render systems inoperable. This paper integrates a dynamic risk-based interdependency model with multi-criteria decision analysis to evaluate dock-specific resource allocation alternatives to provide tangible assessments of port preparedness planning. We analyze a set of discrete allocations for the inland waterway Port of Catoosa in Oklahoma to assess total expected loss per industry.

2 - How Much Should We Spend on Preparing for Disruptions? Cameron MacKenzie, Assistant Professor, Iowa State University, 3004 Black Engineering, Ames, IA, 50011, United States of America, camacken@iastate.edu Every time a disruption causes major economic damages and/or leads to several fatalities, the public questions why more wasn’t done to prepare for the disruption. Prevention and preparedness require allocating resources (usually money) that could be used for other purposes. I will present a model to help a decision maker determine how much should be allocated before a disruption for both a risk neutral and risk averse case. The model is applied to the Deepwater Horizon oil spill. 3 - Infrastucture Resilience Framework for Decision Making Comparison and Real Time Monitoring Dante Gama Dessavre, Ph. D. Candidate, Stevens Institute of Technology, Castle Point on Hudson, Hoboken, NJ, 07030, United States of America, dante.gamadessavre@gmail.com, Jose Emmanuel Ramirez-marquez Understanding impacts of disruptions to systems is a critical component to enable infrastructure modifications that improve their resilience. We present a framework and optimization formulations that enable direct comparison of the resilient behavior of a system due to different design choices; We also show how it can be adapted to monitor the resilient behavior of a system in real time in order to enhance the tools available to decision makers for disruption management. 4 - How Performance Measure Selection Impacts Risk Characterization in Interdependent Infrastructure Allison Reilly, Johns Hopkins University, 3400 N Charles Street, Baltimore, MD, 21218, United States of America, acr@jhu.edu, Andrew Samuel, Seth Guikema A perturbation in one infrastructure system stemming from a failure, an investment, or some other change, is known to impact multiple performance measures in dependent systems. However, most interdependent infrastructure models evaluate influences on performance using only one measure. We evaluate the implications of using one performance metric on decision-making and system-wide risk and discuss how multiple performance measures could be used to build a more holistic view. 5 - Bayesian Method for Thermo-electric Power Generation Drought Risk Assessment Royce Francis, George Washington University, 800 22nd St. NW B1850, Washington, DC, 21212, United States of America, seed@gwu.edu, Behailu Bekera This paper presents a Bayesian framework for thermoelectric drought risk assessment, focusing on current and future risk of drought driven by climate and socio-economic changes. Uncertainties vary across regions based on the climate condition, technology choices, regulations, demographic changes. These factors are taken into consideration in updating uncertainty in parameters of a cost model currently being developed by the authors. WA65 65-Room 113B, CC Decision Analysis Applications in Operations Management Sponsor: Decision Analysis Sponsored Session Chair: Barry Cobb, Missouri State University, 901 S. National Ave., Springfield, MO, 65897, United States of America, BarryCobb@MissouriState.edu 1 - Import Security: Assessing the Risks of Imported Food Jonathan Welburn, PhD Student, University of Wisconsin - Madison, 1513 University Avenue, 3237 Mechanical Engineering, We quantitatively address risks associated with imported foods using data on the rate of FDA import violations, an indicator of risk. Using regression to analyze these risks, we explore how risk varies by product and violation type. Additionally, we explore improvements from finer categorization of product types. While this might make results might even more useful, inconsistencies between current databases make this difficult to achieve. Madison, WI, 53706, United States of America, welburn@wisc.edu, Vicki Bier, Steven Hoerning

394