Informs Annual Meeting Phoenix 2018

INFORMS Phoenix – 2018

SA50

cost, schedule, and performance in terms of a project’s key value attributes and threats to them. It considers how high the “bar is set for a project (its set goals) and therefore how challenging and risky it will be. A project’s capabilities as a “jumper (to clear the bar and meet its goals) determine the portion of its value at risk (VaR). Progress occurs as VaR is reduced: activities “add value by chipping away at the project’s “anti-value, the risks that threaten value. 5 - Capacity Planning for Project Management by Adjustable Distributionally Robust Optimization Daniel Zhuoyu Long, Chinese University of Hong Kong, Room 605, William M. W. Mong Engineering Building, Hong Kong, Antonio J. Conejo, Nicholas G. Hall, Runhao Zhang Many project companies use outsourced providers which require capacity reservation. Capacity reservation decisions provide access to outsourced services that enable project crashing, but must be made before task durations are realized. We model these decisions with the consideration of worst case distribution. Once task durations are realized, the project company makes decisions about fast tracking and outsourced crashing. We model the company’s objective using the target-based measure. We compare the performance of the model against benchmarks from the literature, and show that it provides lower risk and greater robustness to distributional information. n SA52 North Bldg 231C Social Media and Financial Information Emerging Topic: Social Media Analytics Emerging Topic Session Chair: Soo Jeong Hong, PhD, Michigan State University, East Lansing, MI, United States 1 - Social Media Disclosure and Analysts as Information Intermediaries Kwangjin Lee, Michigan State University, 31 Middlevale Road, 1645 f, East Lansing, MI, 48823, United States Social media provide chances of direct communication between firms and investors. On one hand, it may reduce the information intermediary role of financial analysts. On the other hand, information overload increases the demand for financial analysts’ role as information intermediaries. Therefore, it is an empirical question whether prevalence of social media disclosure substitute or complement the traditional intermediaries, financial analysts. I investigate the effect of social media disclosure on the market reaction to forecast revisions and find social media disclosures complement rather than substitute for the intermediary role of financial analysts, consistent with mosaic theory. 2 - Variational Autoencoders and Infinite Edge Partition Models for Text Modeling and Community Detection: Evidence from Yelp.com Review Hyunsang Son, The University of Texas at Austin, Austin, TX, United States Recent works from Natural Language Process (NNP) found that variational autoencoders (VAE) with appropriate decoders such as Long short-term memory (LSTM) better on generative text modeling. By adopting another newly emerged research trends in social network analysis called community detection, this paper will investigate how user networks differ in yelp.com review conditioning on sentiment label. 3 - The Role of Public Sentiment and Firm Crisis Response Strategy on Twitter in Stock Markets During Firm Crisis Situations Soo Jeong Hong, Michigan State University, 404 Wilson Rd, Room 309, East Lansing, MI, 48824, United States, Kwangjin Lee, Jinse You The importance of social media in prediction of stock prices is increasing as the individual investors’ and firm’s social media usage has become prevalent. We seek the associations between sentiment of Twitter messages and stock price behavior during firm crisis situations. This paper is focused on gaining deeper insights into investors’ behavior in order to minimize financial and reputational damages on the firms.

n SA50 North Bldg 231A Joint Session Practice/Practice Curated:Editor’s Cut #1: Cybersecurity, Elections, Transportation, Agribusiness Sponsored: INFORMS Section on Practice (formerly CPMS) Sponsored Session Chair: Carrie Beam, University of Arkansas, Walnut Creek, CA, 94596, United States 1 - Moderator - Editor’s Cut #1 (Cybersecurity, Elections, Transportation, Agribusiness) Robin Lougee, IBM Research, IBM TJ Watson Research Center, 1101 Kitchawan Road, Yorktown Heights, NY, 10598, United States This panel will cover the Cybersecurity, Elections, Transportation, and Agribusiness issues of Editor’s Cut. Come learn more about the case studies, videos, and other journal articles available free, online Panelists Scott Nestler, University of Notre Dame, 51344 Pebble Beach Court, Granger, IN, 46530, United States Alan Erera, Georgia Institute of Technology, School of Industrial & Systems Eng, Atlanta, GA, 30332-0205, United States Saurabh Bansal, Penn State University, 405 Business Building, State College, PA, 16801, United States Sam Ransbotham, Boston College, Fulton Hall 460A, 140 Commonwealth Avenue, Chestnut Hill, MA, 02467, United States n SA51 North Bldg 231B Advances in Project Management Emerging Topic: Project Management and Scheduling, in Memory of Joe Leung, Emerging Topic Session Chair: Janne Kettunen, The George Washington University, Suvrat Dhanorkar, Pennsylvania State University, 466 Business Building, University Park, PA, 16802, United States, Enno Siemsen How does one ensure that tasks get done, even if one has only limited authority and control over them? This is a fundamental problem of project management. Additionally, since tasks do not exist in a vacuum, organizations constantly require their employees to multi-task and switch attention from one task to another. We use energy efficiency projects as the research context and use a combination of archival and experimental data to show that reminders serve as a key managerial lever to focus the employee attention to specific tasks. 2 - Project Portfolio Selection: Disjunctive Stochastic Programming Formulations with Joint Chance Constraint Janne Kettunen, The George Washington University, Department of Decision Sciences, 2201 G. Street NW, Washington, DC, 20052, United States, Miguel Lejeune The main goals of start-up companies in new product development (NPD) are to attain a reliable return level and deliver this return level fast. Achieving these goals is complicated due to uncertainties in projects’ returns and durations. We develop new models for the NPD capturing the above goals. The first model is static representing the waterfall product development process whereas the second model is dynamic representing the agile product development process. We design a reformulation method and a new decomposition algorithm to solve these problems. We apply our models to an US-based software start-up company and derive managerial insights. 3 - Information Exchange in Parallel Search Fang Ni, University of Cologne, Cologne, Germany, Fabian J. Sting This paper studies parallel search for technological innovation by multiple teams, which is a relevant context for organizational innovation. Within such a setting, a critical issue is how to organize information exchange among the searching teams. The study employs an agent-based simulation approach to identify optimal communication strategies for the firm. 4 - Planning, Tracking, and Reducing Project Value at Risk Tyson R. Browning, Texas Christian University, Neeley School of Business, TCU Box 298530, Fort Worth, TX, 76129, United States Uncertainty, risk, and rework challenge projects to meet goals and deliver anticipated value. Conventional techniques for planning and tracking earned value do not account for these. I present a methodology for planning and tracking Washington, DC, 20052, United States 1 - Using Reminders to Focus Attention

20