Informs Annual Meeting 2017

TE26

INFORMS Houston – 2017

2 - A Dynamic Learning Approach to Assortment Selection Vashist Avadhanula, Columbia University, 3022 Broadway, 4th Floor West, New York, NY, 10027, United States, vavadhanula18@gsb.columbia.edu, Shipra Agrawal, Vineet Goyal, Assaf Zeevi We consider a dynamic assortment selection problem where customers choose according to a multinomial logit (MNL) choice model. The retailer observes this choice and the objective is to dynamically learn the model parameters, while optimizing cumulative revenues over a selling horizon of length T. We present an efficient algorithm that simultaneously explores and exploits, achieving performance independent of the underlying parameters. The algorithm can be implemented in a fully online manner, without knowledge of the horizon length T. Furthermore, the algorithm is adaptive in the sense that its performance is near-optimal in both the “well separated” case and the general parameter setting. 3 - The Limit of Rationality in Choice Modeling: formulation, Computation, and Implications Srikanth Jagabathula, NYU. Stern School of Business, 44 W. 4th St, Kmc Rm 8-74, New York, NY, 10012, United States, sjagabat@stern.nyu.edu, Paat Rusmevichientong We focus on quantifying the limit of rationality (LoR) in choice modeling applications, defined as the “cost” of approximating the observed choice fractions from a collection of offer sets with those from the best fitting probability distribution over rankings. Computing LoR is intractable in the worst case. We express the complexity in terms of structural properties of an appropriately defined graph and exploit the graph structure to compute LoR efficiently for a large class of applications. Using real-world grocery sales data, we identify product categories for which going beyond rational choice models is necessary to obtain acceptable performance. 350A Innovation/Entrepreneurship Contributed Session Chair: Pelin Atahan, Ozyegin University, Istanbul, Turkey, pelin.atahan@ozyegin.edu.tr 1 - How do IT Infrastructure and Human Resources Impact Operational Adjustment Agility Peiran Gao, Huazhong University of Science and Technology, 1037 Luoyu Road, Wuhan, China, gaopeiran@hust.edu.cn, Jinlong Zhang, Yeming Gong, Shan Liu This study proposes that two types of IT resource(i.e, IT infrastructure and human resources) enhance operational adjustment agility by facilitating structural IT capabilities, and that the magnitude of facilitation depends on the level of CEO support of IT. We test these relationships based on data from a multi-industry survey. Our results demonstrate that IT infrastructure and human resources do enhance operational adjustment agility through the mediated effects of structural IT capabilities, and that CEO support of IT differentially affects these relationships. 2 - Spatial Patterns and Socioeconomic Dimensions of Shared Accommodations: the Case of Airbnb in Los Angeles Avijit Sarkar, Professor, University of Redlands, 1200 E.Colton Avenue, Redlands, CA, 92373, United States, avijit_sarkar@redlands.edu, Mehrdad Koohikamali, James B. Pick This study examines spatiotemporal aspects and socioeconomic dimensions of shared accommodations within the broader context of the sharing economy. Specifically, we examine how socioeconomic attributes of Airbnb hosts moderated by hosts’ attitudes towards trust and greener consumption influence participation in the sharing economy. Spatial bias in sharing economy participation rates is examined and policy implications for the supply side of shared accommodations are discussed along with generalizability of results. 3 - A Study of Mixed Reality Feature Utilization within Social Networking Applications Alsius David, Doctoral Student in IT, University of North Texas, 1155 Union Circle #311277, Denton, TX, 76203, United States, alsiusdavid@my.unt.edu, Daniel A.Peak Mixed reality (MR), where both real and virtual worlds are mixed, has been trending in social networking applications. MR enhances communication by integrating a virtual experience into the users’ current environment. Taking the user’s perspective, this study identifies whether personality, quality, and environmental characteristics influence feature utilization on the reality-virtuality continuum employed by SnapChat and Facebook Messenger. TE25

4 - The Effect of Emotional Support by Conversational Assistants on Information Disclosure by Users Kambiz Saffarizadeh, Georgia State University, 35 Broad Street, Suite 900, Atlanta, GA, 30303, United States, ksaffarizadeh1@gsu.edu, Maheshwar Boodraj, Tawfiq Alashoor The main objective of this research is to investigate interaction behaviors between CAs and users. Specifically, we will examine how CA-based emotional support can influence users’ satisfaction and disclosure behavior. We will draw upon interpersonal communication theories to develop and test a research framework in an experimental setting using a conversational assistant. We are interested in examining how CA-based emotional support can facilitate higher trust and lower privacy concern toward sharing personal information, which in turn influence the relationship between the user and CA in terms of service satisfaction and users’ disclosure behavior. 5 - A Duopoly Model of Competition in Cloud Computing Services Pelin Atahan, Assistant Professor, Ozyegin University, Nisantepe Mah Orman Sok Cekmekoy, Istanbul, 34794, Turkey, pelin.atahan@ozyegin.edu.tr, Basak Altan, Okan Orsan Ozener In this paper, we investigate the price, quality of service and capacity decisions of ‘Cloud Computing’ service providers. We present a model where customers vary in their valuation of quality and in the quantity they demand. We study a duopolistic market, where the incumbent firm enjoys lower costs in later periods as a result of learning. We study how the service offerings change and how the entrant firm positions itself under two scenarios. In the first case, the incumbent firm does not consider future competition, whereas, in the second case, the incumbent firm anticipates future competition and acts strategically to deter entry in the first period. 350B Cyber Analytics and Optimization Sponsored: Social Media Analytics Sponsored Session Chair: Les Servi, The MITRE Corporation, The MITRE Corporation, Bedford, MA, 01730-1420, United States, lservi@mitre.org 1 - Multifirm Models of Cybersecurity Investment and Competition vs. Cooperation and Network Vulnerability We develop and compare three distinct models for cybersecurity investment in competitive and cooperative situations to safeguard against potential and ongoing threats. We utilize game theory and optimization theory. We then apply the framework to two case studies on retailer and financial services and recommend a course of action. 2 - Scheduling and Training Multi-skilled Analysts under Uncertainty Doug Altner, MITRE Corporation, 7515 Colshire Drive, M/S.H617, McLean, VA, 22102, United States, daltner@mitre.org, Erica Mason, Les Servi This talk introduces a shift scheduling problem with unknown demand, multiple job types, and multi-skilled employees who can be trained to complete additional jobs. We provide a stochastic integer programming formulation of this problem and demonstrate that this formulation can be directly fed into Gurobi to solve instances with as many as 50 employees and 50 scenarios per day without any decomposition techniques. Finally, we present a local search heuristic that consistently produces good solutions for instances that are too large for a direct feed into a commercial solver. 3 - A Methodology to Measure and Monitor Level of Operational Effectiveness of a Cybersecurity Operations Center (CSOC) Rajesh Ganesan, George Mason University, Fairfax, VA, United States, rganesan@gmu.edu, Ankit Shah, Sushil Jajodia, Hasan Cam There are a number of disruptive factors that can adversely impact the normal operating conditions of a cybersecurity operations center (CSOC) such as 1) higher alert generation rates from a few IDSs, 2) new alert patterns that decreases the throughput of the alert analysis process, and 3) analyst absenteeism. The impact of all the above factors is that the alerts wait for a long duration before being analyzed, which impacts the readiness of the CSOC. It is imperative that the readiness of the CSOC be quantified and measures taken to maintain it, which in this talk is defined as the Level of Operational Effectiveness (LOE) of a CSOC. TE26 Anna B. Nagurney, University of Massachusetts Amherst, Isenberg School of Management, Dept of Operations & Information Mgmt, Amherst, MA, 01003, United States, nagurney@isenberg.umass.edu, Shivani G. Shukla

419