Informs Annual Meeting 2017

SB21

INFORMS Houston – 2017

SB20

4 - Queueing Processes on a Random Graph René Bekker, PhD, VU Amsterdam, Netherlands, r.bekker@vu.nl, Nicos Starreveld, Michel Mandjes During this presentation we attenpt to develop a queueing model on a random graph. The queueing process at each node is following the dynamics of an infinite-server queue. The links between nodes (or the nodes themselves) evolve between up and down states according to some background process. Alternatively, the model can be interpreted as a network of infinite-server queues. The performance measures we study are the number of links that are up and the queue content at the nodes. We explicitly determine their moments, give a functional central limit theorem for the number of nodes that are up, and present more detailed results for fully symmetric random graphs. 342A Empirical Research at OM/Finance Interface Sponsored: Manufacturing & Service Oper Mgmt, iFORM Sponsored Session Chair: Volodymyr O. Babich, Georgetown University, Washington, DC, 20057, United States, vob2@georgetown.edu 1 - Supply Chain Implications of Shock Propagation and Systematic Risk Jing Wu, City University of Hong Kong, Hong Kong, jing.wu@cityu.edu.hk, John R.Birge In this talk, by leveraging supply chain data that is unstructured, incomplete, and highly complex, we show empirically the spill-over effects of moderate shocks transmitted to other connected firms in supply chains. We also show the systematic risk implications of the self-organized supply chain structure which cannot be explained by common risk factors. 2 - Stock Market Reaction to the 2015 Volkswagen Diesel Emission Scandal Brian W. Jacobs, Michigan State University, 632 Bogue Street. Rm N370, East Lansing, MI, 48824, United States, jacobsb@bus.msu.edu, Vinod R. Singhal Responsible sourcing is of increasing research interest, but what happens if your customer is not socially responsible? We examine the propagation of changes to shareholder value that emanated from the 2015 VW diesel emissions scandal, spread upstream in the supply chain, and horizontally to competitors and the diesel industry. 3 - Supply Chain Contagion in Credit Default Swap Market Volodymyr O. Babich, Georgetown University, 37th & O.St NW, McDonough School of Business, Washington, DC, 20057, United States, vob2@georgetown.edu, Senay Agca, John R. Birge, Jing Wu We present evidence for the existence of the supply-chain contagion channel in the Credit Default Swap (CDS) market, using event studies and diff-in-diff regressions, based on CDS, supply chain, and firm data from 2004 to 2014. The CDS spread up- (down-) jump in a supply chain is associated with CDS spread increase (decrease) for the reference firm. Supply-chain contagion extends to higher supply-chain tiers. We identify firm and supply chain attributes that contribute to CDS supply chain contagion. For example, older SC links, and links with stronger financial ties between customers and suppliers amplify contagion. We confirm results with robustness tests, e.g., placebo test with inactive links. 4 - Optimal Term Design for Supply Chain Finance Xiaobo Ding, Johnson School, Cornell University, Ithaca, NY, United States, xd72@cornell.edu Small suppliers are often cash-strapped due to long payment collection cycles and high credit rates. Supply chain finance, also known as reverse factoring (RF), is a popular instrument used in practice to address this problem. In this paper, we study the supplier’s cash management problem under RF, and also quantify the value of RF to the supplier. We show that the optimal policy for the multi-period cash management problem is a modified (L,U) policy. We explain how RF creats a win-win-win situation for the supplier, the buyer and the bank. Moreover, our model can be used to inform the design of the optimal term for supply chain finance. SB19

342B Computational Research Methods and Applications in TIMES Sponsored: Technology, Innovation Management

& Entrepreneurship Sponsored Session

Chair: Hyunwoo Park, Ohio State University, 2100 Neil Ave, Columbus, OH, 43210, United States, park.2706@osu.edu 1 - A Bicentric Diagram Approach to Manage Interdependencies when Designing Complex Systems Hyunwoo Park, The Ohio State University, Columbus, OH, United States, park.2706@osu.edu, Manuel Sosa, Rahul C. Basole Complex systems are conceived as network of interconnected components. An important challenge for engineering managers is to identify the components that could potentially be affected if design changes affect two interdependent components. We introduce a novel approach, based on bicentric diagrams, to manage design changes of pairs of interdependent components of complex systems. We illustrate the use of our approach by applying it to the management of various critical technical interfaces during the development of large commercial aircraft engine. 2 - Prediction of Patient Activation During Technology Enabled Continuity of Care Intervention Nitin Joglekar, Boston University, 595 Commonwealth Ave, Boston, MA, 02215, United States, joglekar@bu.edu, Carolyn Queenan, Kellas Ross Cameron Patients’ skills, knowledge, and motivation to actively engage in their healthcare are assessed with the Patient Activation Measure (PAM). We predict PAM as a function of the strength of information signals associated with telemedicine technology enabled care using a machine learning methodology. Predictions are shown to be subject to under/over estimation biases, consistent with the behavioral concept of system neglect in signal detection theory. 3 - Visual Business Ecosystem Intelligence: A Computer Vision Based Approach Rahul C.Basole, Georgia Institute of Technology, School of Interactive Computing, 85 5th Street Nw, Atlanta, GA, 30332, United States, basole@gatech.edu This study presents the application of computer vision, pattern recognition, and data visualization techniques for extracting, analyzing, and understanding image- based data of emerging business ecosystems. 4 - Business Model Innovation Feature Extraction and Application to the Lean Startup Framework Christophe Pennetier, INSEAD, 6 Marina Boulevard, # 27-15, Singapore, 018985, Singapore, christophe.pennetier@insead.edu, Serguei Netessine, Karan Girotra Using a new curated dataset with more than half a million startups, we use state- of-the-art text mining and machine learning techniques to identify business model innovations and study their effects on startups’ success. We are the first to empirically study the lean startup approach. 342C Computer Aided Medical Diagnosis Invited: InvitedHealthcare Systems and Informatics Invited Session Chair: Juri Yanase, Complete Decisions, LLC, Baton Rouge, LA, 70810, United States, jurijuriy@aol.com Co-Chair: Evangelos Triantaphyllou, Louisiana State University, Baton Rouge, LA, 70803, United States, etriantaphyllou@yahoo.com 1 - Translating Prediction to Decision: A Framework to Tackle Risk Identification and Presentation Challenges in Sepsis Muge Capan, Christiana Care Health System, 902 N. Market Street, Apt 527, Wilmington, DE, 19801-3082, United States, Muge.Capan@ChristianaCare.org, Julie Simmons Ivy, Stephen Hoover, Kristen Miller, Jeanne Marie Huddleston, Ryan Arnold Rising incidence and mortality rate make sepsis, a spectrum of acute organ dysfunction secondary to an infection, a significant burden on the healthcare system. While predicting sepsis-induced deterioration using electronic health records promises enhanced risk detection, translating findings into decisions is challenging. We present a framework to reengineer the risk presentation associated with sepsis and in-hospital mortality in septic patients. SB21

41