Informs Annual Meeting 2017

MA27

INFORMS Houston – 2017

3 - The Efficiency Measurement of Colombian Higher Education Institutions using Data Envelopment Analysis with Flexible Measures Sepideh Abolghasem, Universidad de los Andes, Carrera 1 Este #19A-40, Bogota, 11001000, Colombia, ag.sepideh10@uniandes.edu.co, Linas Navas, Felipe Montes, Mehdi Toloo, Andrés Useche, Ricardo Salas, Andrés Medaglia, Roberto Zarama It is vital to assess the efficiency of universities for effective allocation and utilization of educational resources. This work explores the performance efficiency of higher education institutions (HEIs) in Colombia using a data envelopment analysis (DEA) model. We examine the efficiency of HEIs in terms of their achievement, applicability and research using a variant of DEA where we include flexible measures. To measure efficiency, we consider the results of the national exam after completing higher education studies, graduates’ employability rate, the number of published articles, the total number of academic staff, the percentage of PhD and the total number of undergraduate students. The application of DEA enables HEI to identify deficient activities and take appropriate actions accordingly for improvement. 350C Social Media Analytics for Disaster Management Invited: Social Media Analytics Invited Session Chair: Jun Zhuang, University at Buffalo, SUNY, 317 Bell Hall, Buffalo, NY, 14260, United States, jzhuang@buffalo.edu Co-Chair: Samiul Hasan, University of Central Florida, Orlando, FL, 32816, United States, samiul.hasan@gmail.com Co-Chair: Arif Mohaimin Sadri, Purdue University, West Lafayette, IN, 47906, United States, asadri@purdue.edu 1 - Get Insight into Social Media Analysis and Incident Detection Ying Chen, Northwestern University, 600 Foster St, Evanston, IL, 60208, United States, y-chen@northwestern.edu, Archak Mittal, Hani S.Mahmassani The feasibility of using social media data for detecting traffic incidents is evaluated. The presented framework consists of three main components, Twitter data mining, location extraction and traffic management. The approach is implemented using data from the Chicago area; using on-line simulation-based traffic estimation and prediction tools, traffic management strategies developed to reduce the impact of incidents are evaluated. The results confirm the potential of Twitter posts to complement and improve the effectiveness of dynamic traffic management approaches for incident conditions. 2 - Crisis Communication and Rumor Management using Social Media During Disasters Jun Zhuang, University at Buffalo, SUNY, 317 Bell Hall, Buffalo, NY, 14260, United States, jzhuang@buffalo.edu, Bairong Wang Social media has been used by organizations and private citizens for crisis communication during disasters. However, few research has studied the users’ behavior when facing rumors and debunking information. We first study the effectiveness of crisis communication and how re-tweet and mention could help improve crisis information impression. Second, we investigate four cases of rumor responding and corresponding debunking behaviors of Twitter users during Hurricane Sandy in 2012 and Boston Marathon bombings in 2013. Finally, we discuss the optimal debunking strategies dealing with potential rumor information, and the corresponding consequences on the downstream information sharing. 3 - Understanding the Effectiveness of Social Media Communication During Disasters using Twitter Data from Hurricane Sandy Kamol Chandra Roy, University of Central Florida, Orlando, FL, United States, roy.kamol@Knights.ucf.edu, Samiul Hasan Rapid communication during a natural disaster is one of the critical aspects of effective disaster management strategies. Online social media provides a tremendous opportunity to communicate information during disasters. In this study, around 52.5 million tweets related to hurricane Sandy have been analyzed to assess the effectiveness of information spreading in social media. Our study reveals that information spreading capacity depends on user follower count, user category, bot score (controlled by human or machine) and activity levels. MA27

4 - Analysis of User-Specific Communication Patterns of Hurricane Sandy using Twitter Data Arif Mohaimin Sadri, Purdue University, 149 Arnold Drive Apt 12, West Lafayette, IN, 47906, United States, asadri@purdue.edu Hurricane Sandy was one of the strongest and costliest in the history of hurricanes. Many people used social media to communicate during this period while lacking access to traditional information sources. In this study, we analyzed raw data (~52 M tweets, ~13 M users, Oct 14 -Nov 12, 2012) obtained from Twitter. First, we identify different communities from the subgraphs of Twitter that was active before, during, and after Sandy’s landfall. Then, we extract different crisis communication topics evolved in these subgraphs while Hurricane Sandy was approaching. 5 - Revisiting Collaborative Activity and its Impact on Peer-produced Content Srikar Velichety, University of Memphis, 395 S.Highland Street, Suite 128, Apt#301, Memphis, TN, 38111, United States, svlchety@memphis.edu We provide a bottom-up view of collaboration in peer-produced content. Using the dictionary definition of collaboration, we identify aspects of collaboration that the present research on online communities does not consider. We introduce and define the concept of implicit collaboration and then identify two dimensions and four possible areas of collaboration. In each area, we identify the relevant social network that captures collaboration. Using customized metrics on these networks that capture various aspects of collaboration in each of the four different areas, we quantify the impact of collaboration on article quality. We also measure the usefulness of these metrics in predicting quality. Invited: Auctions Invited Session Chair: Elena Katok, University of Texas at Dallas, Richardson, TX, 75080, United States, ekatok@utdallas.edu Co-Chair: Cuihong Li, University of Connecticut, Storrs, CT, 06269, United States, cuihong.li@uconn.edu 1 - The Reference Price Effect of Reserve and Bin Prices Ernan E.Haruvy, University of Texas-Dallas, School of Management, SM.34 Box 830688, Richardson, TX, 75083-0688, United States, eharuvy@utdallas.edu Reserve prices and BIN prices can serve as reference prices for bidders and in turn influence bidder entry and willingness to pay. We first quantify the dual impact and implied trade-off of price floors and BINs on the number of bidders and auction revenue. Second, we distinguish the reference price effect of reserve prices from the better known truncation and entry effects of reserve prices. Third, we examine robustness of the reserve price effect to the introduction of BINs and seek to characterize the relative reference pull of a reserve price relative to a BIN on the final price. We find that reserve prices have a smaller impact than BIN on final price and that the impact of reserve price is robust to BINs. 2 - The Influence of Social Media on Charitable Fundraising Peter T. Popkowski Leszczyc, University of Alberta, 4-30 F Faculty of Business Blg, Edmonton, AB, T6G 2R6, Canada, ppopkows@ualberta.ca We investigate the combined power of auctions and social media to raise funds for charity. Auctions are uniquely positioned to price unique art, and social media is uniquely positioned to generate demand for that art. We auctioned off artwork created by teams of under-privileged youth with unique stories and goals, and used social media, with a focus on Facebook and Facebook likes to disseminate word of mouth about these auctions. We find that social media increases prices in charity auctions by affecting (1) impressions (i.e., the number of times a Facebook posting displayed) which in turn affect bidding intensity and then price and (2) through Facebook likes which affect price directly. 3 - Procurement Mechanisms with Post-auction Cost-reduction Investigations George Chen, University of Texas at Dallas, Richardson, TX, United States, georgeqc@umich.edu, Damian Beil, Izak Duenyas Consider the following procurement setting: After seeing suppliers’ bids, the buyer can choose to investigate some suppliers to identify cost-reduction opportunities; the investigated suppliers’ costs will decrease by the identified cost- reductions; the buyer then awards the contract. We characterize the optimal first-price sealed-bid mechanism, as well as an optimal mechanism among a wider class of mechanisms. For ex-ante symmetric suppliers, incorporating investigations create an ex ante win-win situation for everyone. The win-win may break down when suppliers are sufficiently asymmetric, but no supplier has the incentive to unilaterally block investigation. MA28 350D Auctions and Market Design

142