Informs Annual Meeting 2017

TC28

INFORMS Houston – 2017

2 - A Computational Study for the Unit Commitment Model Jianqiu Huang, University of Florida, 411 Weil Hall, Dept of Industrial and System Engineering, Gainesville, FL, 32608, United States, jianqiuhuang@ufl.edu, Kai Pan, Yongpei Guan In this talk, we present the recent progress on the polyhedral study of the power system operation problems. The computational studies verify the effectiveness of our proposed approach. 3 - Unified Formulations for Combined-Cycle Units Yongpei Guan, University of Florida, 303 Weil Hall, P.O. Box 116595, Gainesville, FL, 32611, United States, guan@ise.ufl.edu, Lei Fan, Yanan Yu In this talk, we describe and develop edge-based and node-based formulations for combined-cycle units (CCUs) with different levels of accuracy in respecting physical restrictions of each turbine, from no restriction requirement of each turbine, to include min-up/-down time restrictions of each turbine, to finally include most restrictions such as min-up/-down time, capacity, and ramping restrictions of each turbine. We develop corresponding models for Independent System Operators (ISOs) and Vertically Integrated Utilities (VIUs), respectively. These models demonstrate the complexity in terms of the size of decision variables and show the tradeoff between accuracy and computational complexity. In general, to the best of our knowledge, this letter provides one of the first models that can capture both configuration-based model needs and all physical restrictions for each turbine within CCUs. 4 - Improving International Oil and Gas Supply Projections Tuncay Alparslan, Operations Research Analyst, U.S. Department of Energy, Washington, DC, United States, Tuncay.Alparslan@eia.gov While there are many models of energy production, no existing model adequately estimates worldwide oil and gas production, conversion, and logistics in a single integrated decision framework that also reflects coproduction of oil and gas resources. To better capture the interplay between shared resources and markets, U.S. Energy Information Administration is developing the Global Hydrocarbon Supply Module (GHySMo). GHySMo provides flexible geographic resolution, endogenous modeling of other fuels, and resource trade-offs that will advance current models. We will present model structure, core innovations, and early results with a focus on logistics and integration. 350B Network Analytics and Telecommunications Sponsored: Telecommunications Sponsored Session Chair: Michael Bartolacci, Penn State Berks, Penn State Berks, bethlehem, PA, 18017, United States, mbartolacc@aol.com 1 - A Dynamic Flocking Algorithm for Topology Control in Mobile Ad HOC Networks Abdullah Konak, Penn State Berks, Tulpehocken Road, PO.7009, Reading, PA, 19610, United States, konak@psu.edu Mobile Ad Hoc Networks (MANETs) has many application areas, including sensor networks, vehicular networks, robot swarms, and disaster emergency networks. On the other hand, MANETs have dynamic topologies which could be disconnected because of mobility of nodes. This paper presents an improved flocking algorithm to maintain the connectivity of a MANET using autonomous and intelligent agents. In the proposed flocking algorithm, agents modify their behavior parameters dynamically based on the size of their neighborhoods. A simulation study is conducted to investigate the performance of the new algorithm. Simulation results show that the algorithm outperforms earlier approaches. 2 - Cooperative Human-centric Sensing (HCS) Communications Albena Mihovska, Aalborg University, Center for TeleInfrastruktur, Fredrik Bajers Vej 7C1, Aalborg, 9000, Denmark, amihovska@btech.au.dk Human-Centric Sensing (HCS) is a newly emerged concept in the context of the Internet of Things (IoT) and the active and assisted living (AAL) scenario. HCS connectivity, also referred to as “smart connectivity” enables applications that are highly personalized and often time-critical. In a typical HCS scenario, there may be many hundreds of sensor streams connections; centered around the human, who would be the determining factor for the number, the purpose, the direction and the frequency of the sensor streams. This conribution investigates the delivery of a personalized assistive applications for a scenario of several users within the same HCS environment. 3 - A Distributed Server Provisioning Algorithm for Data Centers with Nonstationary User Requests Yongkyu Cho, POSTECH, 77 Cheongam-ro, Pohang-si, Korea, Republic of, yongkyu.cho@lstlab.org, Young Myoung Ko In the era of growing information communication technology, large-scale data centers are inevitable solutions for processing and storing huge amount of TC26

incoming data. Large-scale data centers, however, consume tremendous amount of energy. Achieving energy-efficiency is one of the key problems in data center operations. In the previous research study, we proposed a distributed speed scaling and load balancing algorithm for reducing energy consumption while attaining a desired quality of service when the user requests are stationary stochastic processes. In this research study, we seek how to extend the previous algorithm when the user requests are arriving in nonstationary fashion.

TC27

350C Social Media Analysis: Marketing and Management Applications Invited: Social Media Analytics Invited Session Chair: Soo Jeong Hong, Michigan State University, East Lansing, MI, 48824, United States, hongsoo3@msu.edu 1 - The Effect of Social Media Information on Financial Analysts Kwangjin Lee, Michigan State University, East Lansing, MI, 48824, United States, leekwan6@msu.edu The news media is an important information intermediary in financial markets. Social media encourage participation, collaboration, and information sharing and has transformed the disclosure landscape. While social media allow investors to become more informed and thereby contribute to a reduction in information asymmetry, there is limited evidence on how social media messages affect financial analysts. Using a sample of S&P 500 firms over the period 2012 - 2014, I examine whether and how social media information influences financial analysts. 2 - Effect of Social Media Messages on Stock Price Behavior During Firm Crises Soo Jeong Hong, Michigan State University, 404 Wilson Rd, Room 309, East Lansing, MI, 48824, United States, hongsoo3@msu.edu This study examines whether and how social media messages influence on stock price behavior under firm crises situations. Regression analyses on a sample of S&P 500 firms from 2004 to 2015 show the effect of volume and valence of social media messages on abnormal returns. Results suggest theoretical and practical implications. Invited: Auctions Invited Session Chair: Benjamin Brooks, University of Chicago, Chicago, IL, United States, babrooks@uchicago.edu 1 - An Ascending Auction with Multidimensional Signals Tibor Heumann, tahe@princeton.edu A single-unit ascending auction in which agents observe multidimensional Gaussian signals about their valuation of the good is studied. A class of equilibria is constructed in two steps: (i) the private signals of each agent are projected into a one-dimensional equilibrium statistic, and (ii) the equilibrium strategies are constructed “as if” each agent observed only his equilibrium statistic. Novel predictions of ascending auctions that arise only when agents observe multidimensional signals are provided. 2 - Selling to Intermediaries: Optimal Auction Design in a Common Value Model We characterize revenue maximizing auctions when the bidders are intermediaries who wish to resell the good. The bidders have differential information about their common resale opportunities: each bidder privately observes an independent draw of a resale opportunity, and the highest signal is a sufficient statistic for the value of winning the good. If the good must be sold, then the optimal mechanism is simply a posted price at which all bidders are willing to purchase the good, and all bidders are equally likely to be allocated the good, irrespective of their signals. If the seller can keep the good, then under the optimal mechanism, all bidders make the same expected payment and have the same expected probability of receiving the good, independent of the signal. Conditional on the good being sold, the allocation discriminates in favor of bidders with lower signals. In some cases, the optimal mechanism again reduces to a posted price. The model provides a foundation for posted prices in multi- agent screening problems. Benjamin Brooks, University of Chicago, Chicago, IL, United States, Dirk P.Bergemann, Stephen Morris TC28 350D Auctions and Economics

345