2015 Informs Annual Meeting

SC28

INFORMS Philadelphia – 2015

SC28 28-Room 405, Marriott

these topics in a session with some of the leading thinkers in both OR and Analytics. Be sure to attend to have your questions answered on these highly complementary and valuable fields.

Market Design Cluster: Auctions Invited Session Chair: Gabriel Weintraub, Professor, Columbia University, Uris Hall, New York, NY, 10027, United States of America, gyw2105@columbia.edu 1 - Incentive Issues in Paired Organ Donation Eduardo Azevedo, Assistant Professor, Wharton, 3620 Locust Walk, Wharton, SHDH 1400, Philadelphia, PA, 19102, United States of America, eazevedo@wharton.upenn.edu, Nikhil Agarwal, Itai Ashlagi, Clayton Featherstone In the last few years a new type of organ donation has arisen. In a paired kidney exchange two recipients with incompatible live donors receive organs from each other’s live donor. Sometimes transactions involve more recipients and/or donors. While many exchanges happen in a decentralized way, others happen in large centralized exchanges. We empirically examine how agents in these markets respond to incentives and whether incentives are misaligned with social goals. 2 - Optimal Procurement Mechanisms for Differentiated Products Gabriel Weintraub, Professor, Columbia University, Uris Hall, New York, NY, 10027, United States of America, gyw2105@columbia.edu, Daniela Saban We study the mechanism design problem faced by a buyer that selects an assortment of differentiated products and unit prices from a set of suppliers with private costs. Then, consumers can choose their most preferred product from this set. The buyer maximizes consumer surplus; to do so, he must balance the trade- off between variety and price competition. We characterize the optimal mechanism and use these results to analyze practical mechanisms. 3 - Efficiency and Stability in Large Matching Markets Yeon-koo Che, Columbia University, 420 West 118th Street, 1029 IAB, New York, NY, United States of America, yeonkooche@gmail.com, Olivier Tercieux We study efficient and stable mechanisms in matching markets when the number of agents is large and individuals’ preferences and priorities are drawn randomly. When agents’ preferences are correlated over objects, the prevailing mechanisms are either inefficient or unstable even in the asymptotic sense. We propose a variant of deferred acceptance which is asymptotically efficient, asymptotically stable and also asymptotically incentive compatible. 4 - Market Fragmentation Rakesh Vohra, University of Pennsylvania, 3718 Locust Walk, Centralized markets reduce the costs of search for buyers and sellers. Their `thickness’ increases the chance of of order execution at competitive prices. In spite of the incentives to consolidate, some markets, have fragmented into multiple trading venues. We argue in this paper that fragmentation is an unavoidable feature of any centralized exchange. Our argument introduces a new way to think about participation in a mechanism when the outside option is endogenous. SC29 29-Room 406, Marriott Joint Session Analytics/CPMS: Panel Discussion on OR vs Analytics: Experts Weigh in on the Case for Change Sponsor: Analytics/CPMS Sponsored Session Chair: Glenn Wegryn, Principal, Analytic Impact LLC, 7555 De Mar Rd, Cincinnati, OH, 45243, United States of America, gwegryn@gmail.com 1 - Panel Discussion on OR vs Analytics: Experts Weigh in on the Case for Change Moderator: Glenn Wegryn, Principal, Analytic Impact LLC, 7555 De Mar Rd, Cincinnati, OH, 45243, United States of America, gwegryn@gmail.com, Panelists: Robert Rose, Anne Robinson, Jack Levis, Don Kleinmuntz, Thornton May, David Dittmann, Jeffrey Camm Operations Research? Management Science? Analytics? What’s in a brand name? How has the emerging field of Analytics impacted the Operations Research Profession? Is Analytics part of OR or the other way around? Is it good, bad, relevant, a nuisance or an opportunity for the OR profession? Is OR just Prescriptive or is it something more? In this panel discussion, we will explore Philadelphia, PA, United States of America, rvohra@seas.upenn.edu, Ahmad Peivandi

SC30 30-Room 407, Marriott Creating and Sustaining an Analytics Advantage Sponsor: CPMS Sponsored Session Chair: Peter Bell, Ivey Business School at Western University, 1255 Western Road, London, ON, N6G 0N1, Canada, pbell@ivey.uwo.ca 1 - Advanced Analytics at Verizon Wireless Supply Chain Anne Robinson, Verizon Wireless, anne.robinson@verizonwireless.com, Mustafa Dogru Supply Chain Advanced Analytics Team carries out projects in the areas of forecasting, inventory optimization, assortment planning, etc. We give an overview of our engagement approach with the internal business units, discuss how our models are implemented and transferred to the business, and conclude with the lessons learned. 2 - Capability Portfolio Analyses Tool (CPAT) The Capability Portfolio Analyses Tool (CPAT) delivers unparalleled insight into multi-decadal modernization planning for the Army’s large ground combat fleet. This first-of-its-kind capability optimizes modernization plans while considering the complex interrelationship between performance, budgets, schedules, production capacities, etc. Over 40 studies have been completed- informing stakeholders up to Army Vice Chief of Staff, OSD, and Army Acquisition Executive levels. 3 - The Analytics Advantage: Some Reflections from Business and Policy-making Doug Samuelson, InfoLogix, Inc., 8711 Chippendale Court, Annandale, VA, 22003, United States of America, samuelsondoug@yahoo.com Analytics / OR / MS (or whatever we want to call it) involves injecting more intellectual rigor into decision-making. Key elements of how to do this include using the decision-makers’ language, understanding their needs and time frame, and building multiple constituencies of decision-makers who trust your recommendations. This is not as obvious in practice as it sounds in summary. 4 - Sustaining an Analytics Advantage Many claim that the use of analytics can create a competitive advantage but research suggests that any advantage created is often fleeting: most analytics is easily replicated. There are, however, examples of firms that have sustained a competitive advantage through analytics for years, even decades. How have they done this? What are the lessons for analytics leaders? This presentation will provide examples that may provide lessons on how to sustain an advantage created through analytics. SC31 31-Room 408, Marriott Interactions Between Learning and Decision Making Sponsor: Data Mining Sponsored Session Chair: Theja Tulabandhula, Xerox Research Centre India, Bangalore, Bangalore, India, theja2t@gmail.com 1 - The Forest or the Trees? Tackling Simpson’s Paradox with Classification and Regression Trees Galit Shmueli, Professor, National Tsing Hua University, Institute of Service Science, 101, Sec. 2, Kuang-Fu Road, Hsinchu, 30013, Taiwan - ROC, galit.shmueli@gmail.com, Inbal Yahav Social science research often uses observational data to study causality because randomized experiments are impossible, unethical or expensive. Simpson’s paradox arises in observational causal studies, due to confounders. We propose using Classification & Regression Trees to detect Simpson’s paradox in large observational data. The approach relies on the tree structure and the location of the cause vs. the confounders in the tree. We illustrate our approach on several real applications. Pooja Dewan, BNSF Railway, Fort Worth TX 76092, United States of America, Pooja.Dewan@bnsf.com Peter Bell, Ivey Business School at Western University, 1255 Western Road, London, ON, N6G 0N1, Canada, pbell@ivey.uwo.ca

102