2015 Informs Annual Meeting

MC28

INFORMS Philadelphia – 2015

MC29 29-Room 406, Marriott

4 - Bi-objective Route Planning for Unmanned Air Vehicles in Continuous Space Diclehan Tezcaner Ozturk, Dr., TED University, Industrial Engineering, Ankara, Turkey, diclehan.ozturk@tedu.edu.tr, Murat Koksalan We consider the route planning problem of a single unmanned air vehicle visiting multiple targets in continuous space under two objectives: minimization of total distance traveled and minimization of total radar detection threat. We develop solution approaches to generate the nondominated frontier of this problem. We also develop an interactive algorithm that asks for comparison between solutions and progressively reduces the search area around the decision maker’s most preferred solutions.

Joint Session Analytics/CPMS: Panel Session: Extending the Reach of Certification: The New Associate Certified Analytics Professional

Sponsor: Analytics Sponsored Session

Chair: Polly Mitchell-Guthrie, Sr. Mgr., Advanced Analytics Customer Liaison Group, SAS Institute, SAS Campus Dr., Cary, NC, 27513, United States of America, Polly.Mitchell-Guthrie@sas.com 1 - Extending the Reach of Certification: The New Associate Certified Analytics Professional Moderator:Polly Mitchell-Guthrie, Sr. Mgr., Advanced Analytics Customer Liaison Group, SAS Institute, SAS Campus Dr., Cary, NC, 27513, United States of America, Polly.Mitchell- Guthrie@sas.com, Panelists: Louise Wehrle, Esma Gel, Randy Bartlett The Certified Analytics Professional (CAP®) Program will soon be within reach of more people with the launch this fall of the Associate Certified Analytics Professional. This new certification will retain the rigor of the current CAP and will allow candidates to pass the certification exam without the required work experience if those candidates have an MS in a related field. To earn the CAP, experience is still required. We expect this new offering to be particularly attractive to graduates of the many MS in Analytics degree programs where demand has been high. This panel will include the Chair of the Analytics Certification Board; the Certification Program Manager; a member of the Analytics Certification Board who is also the INFORMS VP of Sections and Societies; and a noted author, speaker, analytics professional as well as CAP. The panel will discuss the overall CAP program, the new offering, and where the program is headed. MC30 30-Room 407, Marriott Practice Presentations by INFORMS Roundtable Companies II Sponsor: INFORMS Practice Sponsored Session Chair: Stefan Karisch, Digital Aviation Optimization & Value Strategy, Boeing Commercial Aviation Services, 55 Inverness Drive East, Englewood, CO, 80112, United States of America, stefan.karisch@jeppesen.com 1 - A Consultancy’s Perspective on the Evolution of and Future Opportunity for Operations Research Nick Nahas, Booz Allen Hamilton, 8283 Greensboro Drive, McLean, VA, 22102, United States of America, nahas_nicholas@bah.com, Cenk Tunasar Big Data is popular and computation is fast and distributed, yet has the practice truly taken advantage of this in the same way as the Data Scientist? And what can the practice do, to use Big Data to enhance management decision making? And what innovations will this lead to? Booz Allen will explore these questions from our perspective as a global management and technology consulting firm with a diverse set of experiences solving today’s greatest challenges in the public and private sector. 2 - Change Management for Analytics Projects Irv Lustig, Optimization Principal, Princeton Consultants, 2 Research Way, Princeton, NJ, 08540, United States of America, irv@princeton.com, Zahir Balaporia, John Milne, Karl Kempf, Rahul Saxena To gain the acceptance of analytics used for decision support, analytics professionals should use organizational change management principles. This paper provides advice to these professionals on these principles as they work with key users to gain acceptance of analytics. These ideas are drawn from the collective observations and insights of practitioners with decades of experience in the development and deployment of analytic methods for improved decision making.

MC28 28-Room 405, Marriott Auctions and Trading Agents Cluster: Auctions Invited Session

Chair: Wolf Ketter, Professor, Rotterdam School of Management, Burgemeester Oudlaan 50, Rotterdam, 3062 PA, Netherlands, WKetter@rsm.nl 1 - Understanding, Replicating, and Leveraging Dynamics of Bidder Behavior in Combinatorial Auctions Ali Mahdavi, Carlson School of Management, 321 19th Ave S, #3-365, Minneapolis, MN, 55455, United States of America, mahd0008@umn.edu We use an agent-based modeling approach to simulate human bidder behaviors observed in continuous combinatorial auctions. After validating our bidding agents, we leverage them to simulate a wide variety of competition types (i.e., compositions of bidder behavior) and demonstrate how different competition types affect auction outcomes, such as revenue and allocative efficiency. 2 - An Experimental Agent-based Approach for Intelligent Decision Making on Electricity Future Markets Derck Koolen, Rotterdam School of Management, Erasmus University, Burgemeester Oudlaan 50, Rotterdam, 3062 PA, Netherlands, koolen@rsm.nl, Wolf Ketter, Liangfei Qiu, Ronald Huisman We study the determinants of forward prices and hedging decisions on electricity future markets in an experimental agent-based setting. Faced with imperfect storability and depending on price expectations, the agent supports producers using different technologies with varying risks from fuel trading, emission prices and weather conditions. Presenting different information treatments, the decision making of wholesale market participants is assessed. 3 - Moral Hazard in Auctions: A Principal-agent Model of Bidding Firms We model auctions where the bidders are firms consisting of a principal and an agent. The agent wants to win the package with the highest expected value. Bayesian Nash equilibrium strategies in sealed-bid auctions illustrate possibilities for profitable manipulation of agents. For small markets we prove a non-truthful ex-post Nash equilibrium. Sometimes it is impossible for the principal to set allowances such that the agent bids truthful. In larger markets, there are multiple equilibria. 4 - An Auction Mechanism for Scheduling Electric Vehicle Charging Konstantina Valogianni, Rotterdam School of Management, Erasmus University, Burgemeester Oudlaan 50, Rotterdam, Netherlands, kvalogianni@rsm.nl, Wolfgang Ketter, Soumya Sen, Alok Gupta, Eric Van Heck We present an auction-based framework for scheduling electric vehicle charging. We account for individual arrival and departure preferences and aim to service as many customers as possible without suffering major delays regarding their departure time. We show how the payments need to be allocated to minimize the system’s delay cost. We examine the effect of the presented mechanism to smart grid’s stability and reliability. 5 - Dynamic Decision-making in Sequential B2b Auctions Yixin Lu, Assistant Professor, VU University Amsterdam, Netherlands, y2lu@vu.nl, Eric Van Heck, Alok Gupta, Wolfgang Ketter We develop a dynamic structural model of competitive bidding in sequential B2B auctions. Given that bidders often have multiple purchase opportunities in these auctions, we formulate the optimal bidding problem as a partially observable Markov decision process. We apply the model to a unique dataset from the world’s largest flower wholesale market. Martin Bichler, Professor, TU München, Boltzmannstr. 3 Germany, bichler@in.tum.de, Per Paulsen, Salman Fadaei

214