2015 Informs Annual Meeting

WA39

INFORMS Philadelphia – 2015

WA36 36-Room 413, Marriott Joint Session PPSN/TSL: Network Infrastructure Recovery and Resilience Sponsor: Public Sector OR and TSL Sponsored Session Chair: Ozlem Ergun, Assoc. Prof, Northeastern University, 360 Huntington Avenue, Boston, ma, 02115, United States of America, o.ergun@neu.edu 1 - Restoration of Network Connectivity in Large-scale Disaster Response Problems Aybike Ulusan, PhD Student, Northeastern University, 360 Huntington Avenue, Boston, MA, 02115, United States of America, ulusan.a@husky.neu.edu, Ozlem Ergun The goal of this study is to establish the connectivity of a disrupted road network in a timely manner by determining a schedule for recovering/repairing damaged edges. Due to the complex nature of the problem, we propose a heuristic that prioritizes the edges based on their head and tail nodes’ centrality measures. By capturing the particular features of the network topology with these measures, we are able to acquire near-optimal solutions in a rapid fashion. 2 - On the Value of Information-sharing in Interdependent Infrastructure Restoration Thomas Sharkey, Rensselaer Polytechnic Institute, 110 8th Street, Troy, NY, 12180, United States of America, sharkt@rpi.edu, Huy Nguyen, John Mitchell, William Wallace We consider the problem of restoring multiple disrupted infrastructure networks after an extreme event. This work analyzes the loss in restoration effectiveness resulting from decentralized planning across these networks in forming their restoration plans. We then examine different levels of information-sharing schemes and their impact in reducing this loss. Computational results based on realistic damage scenarios to networks are presented. 3 - An Integrated Network Design and Scheduling Problem for Network Restoration Suzan Afacan, Graduate Student, University of Wisconson

3 - A Framework for Pediatrics Clinic No-show Prediction using Elastic Net and Bayesian Belief Network Kazim Topuz, Wichita State University, 3737 N. Rushwood St. Apt. 1205, Wichita, KS, 67226, United States of America, mktopuz@gmail.com This study predicts the no-show probability of the patient with using demographic, social and appointment as well as appointment attendance information. We develop a hybrid probabilistic prediction framework based on statistics and Bayesian network. We utilized Elastic Net for selection of the variables and state of art structural learning algorithm for building the Bayesian Belief Network. WA39 39-Room 100, CC Supply Chain Management with Marketing Considerations Cluster: Operations/Marketing Interface Invited Session Chair: Yusen Xia, Georgia State University, 35 Broad St., Atlanta, GA, United States of America, ysxia@gsu.edu 1 - Product Quality Strategy in the Presence of Consumer Heterogeneity of Technology Platform Xiuli He, University of North Carolina Charlotte, 9201 University City Blvd, Charlotte, NC, United States of America, xhe8@uncc.edu, Yong Zha This paper investigates quality strategy of the technology product which is attached to the platform when consumers convey a transaction utility on platform. We show that seller will choose different product quality strategies when consumers are homogeneous or heterogeneous on platform and seller has different platform cost strategies. 2 - Modeling Risk and Ambiguity-on-Nature in Normal-Form Games Jian Yang, Associate Professor, Rutgers University, 1 Washington Park RM1084, Newark, NJ, 07102, United States of America, jyang@business.rutgers.edu We propose multi-player frameworks that mitigate decision-theoretical difficulties with the traditional normal-form game. We react to Allais’s (1953) paradox by concerning players with potentially nonlinear functionals of the payoff distributions they encounter. To deal with Ellsberg’s (1961) paradox, we let players optimize on vectors of payoff distributions in which every component corresponds to one nature action. Equilibria are linked to nonlinear programs. 3 - Supply Chain Excellence and Firm Values Min Shi, Associate Professor, California State University, Los Angeles, 5151 University Drive, Los Angeles, CA, 90032, United States of America, mshi2@exchange.calstatela.edu, Wei Yu We empirically investigate the impacts of supply chain excellence, measured by AMR Research’s Supply Chain Top 25 list, on firms’ financial market performance under different macroeconomic scenarios from 2004-2014. In addition, this paper examines how the characteristics of SCM excellence influence the leading SCM firms on the financial market. 4 - Price or Quality-based Competition and Channel Structure with Consumers Loyalty Sujuan Wang, Institute of Management and Decision, Shanxi University, NO. 92 Wucheng Road, Xiaodian District, Taiyuan, 030006, China, tt_kkcn@sxu.edu.cn, Qiying Hu In the practices of Chinese household appliance industry, we introduce a model for two manufacturers with different customer loyalty who may sell products through a decentralized channel or an integrated channel to consumers and face deterministic demand depending on quality, retail price and customer loyalty. We show that channel choices are closely related to market types and find decentralized channel structures is more likely.

Madison, 1415 Engineering Dr, Madison, WI, 53706, United States of America, iloglu@wisc.edu, Laura Mclay

Infrastructure recovery is important for delivering time-sensitive services and commodities after a disaster. To examine this issue, we present an extension of the p-median problem that allows for network components to be installed by repair crews with a goal of minimizing the cumulative weighted distance. The model is illustrated with a computational example.

WA38 38-Room 415, Marriott Bayesian Approach I Contributed Session

Chair: Babak Zafari, The George Washington University School of Business, 2201 G Street NW, Funger Hall, Suite 415, Washington, DC, United States of America, zafari@gwu.edu 1 - Bayesian Inference for Deadline Reactivity Ji-eun Kim, PhD Student, The Pennsylvania State University, Department of Industrial Engineering, 310 Leonhard Building, University Park, PA, 16802, United States of America, jxk594@psu.edu, David A. Nembhard In time management phenomenon, human generally exhibit the rush before a deadline. This means relatively little time is devoted to tasks at the beginning, and most of the work is performed in close time proximity to the deadline. A Bayesian inference is applied to course website data to obtain reliable individual differences in pacing styles. We observe large reductions in error on data sets, which suggests that Bayesian estimation is a useful tool for learning deadline reactivity. 2 - Bayesian Estimation of Time of the First Bid in Retail Secondary Market Online Auctions Babak Zafari, The George Washington University School of Business, 2201 G Street NW, Funger Hall, Suite 415, Washington, DC, United States of America, zafari@gwu.edu In this work, we propose to develop a model to estimate the distribution of the time of the first bid in retail secondary market online auctions. The proposed estimation is based on a Bayesian mixture model of finite beta distributions. Our main interest is to study this distribution from auctions heterogeneity point of view. We also discuss managerial implications and suggest how auctioneers can benefit from this study.

385