Informs Annual Meeting 2017

TE53

INFORMS Houston – 2017

2 - Technological Discontinuity and Market Competition in the Mobile Service Industry Xiahua Wei, University of Washington, Bothell School of Business, 18115 Campus Way NE, Box 358584, Bothell, WA, 98011, United States, xhwei@uw.edu We study how technological discontinuity contributes to market competition in the mobile service industry. Based on a firm-level panel dataset, we examine whether the ability of new entrants to disrupt incumbents depends on the responsiveness of incumbents to new technologies, and whether technological discontinuities by incumbents can delay industry shakeouts. 3 - Cross-culture Collaboration and Research Citation Impact Xiaopeng Wang, Xi’an Jiaotong University, Xi’an, Shaanxi, 710049, China, Archy714@126.com, Jun Lin, Cui Wentian This research investigated how national cultural difference among scientific teams affect research citation impact. We classified national culture into 5 culture clusters base on Hofstede’s five cultural indicators. The results from the bibliometric analysis showed that articles written by authors from at least two culture clusters has a higher citation impact. Furthermore, we demonstrated a positive moderating effect of focal author’s prestige and a negative moderating effect of discipline difference on the relation mentioned above. We also found that increasing the culture diversity (amount of culture clusters involved) of teams has an inverted-U shaped relation with impact. 4 - Optimal Two-period Response Strategy To New Technology Xiaoxi Liu, Korea University, Seoul, Korea, Republic of, polar1018@korea.ac.kr, Byungcho Kim Technology develops so rapidly, thus when a new technology appears, innovative firms face a dilemma of whether to keep an old technology along with a new one, or drop it. In reality, many industries show the co-existence of both technologies despite the cost of maintaining two product lines. This paper introduces a two- period decision model that examines a firm’s response strategy to the emergence of new technology. In the first period, the firm decides whether to keep the old product line, and in the second period, it chooses whether to improve old version’s quality. The equilibrium price and quality levels are investigated and the This study investigates the development and longitudinal trends in human reliability analysis (HRA) during 1986-2015 through the bibliometric analysis on a data set of 4695 publication records retrieved from Web of Science and Ei Compendex. The number of the HRA documents is used as the performance indicator and examined through both the temporal and geographical perspectives. To reveal the conceptual evolutions of HRA research, the three decades are divided into four consecutive periods (1986-1999, 2000-2006, 2007-2011, 2012- 2015), and the research fronts in each period are investigated through SciMAT. 6 - Entering Convergent Product Market Dawoon Jung, PhD Candidate, Korea University, Business School, LG-Posco Hall, # 408, Seoul, 02841, Korea, Republic of, bogsil@korea.ac.kr, Byung Cho Kim, Bosung Kim Technological convergence is an attempt to integrate disparate technologies across the conventional boundaries and its outcome is convergent product which often exhibits irreplaceable synergetic advantages. This paper examines a firm’s strategic choice that confronts the technological convergence with the view of targeting the growing demand and the synergy from convergence. 361E Robust and Data-Driven Optimization Sponsored: Optimization, Optimization Under Uncertainty Sponsored Session Chair: Phebe Vayanos, University of Southern California, Los Angeles, CA, 90089, United States, phebe.vayanos@usc.edu 1 - Small Data Linear Optimization Vishal Gupta, University of Southern California, Marshall School of Business, 3670 Trousdale Parkway, Los Angeles, CA, 90026, United States, guptavis@usc.edu, Paat Rusmevichientong Optimization problems frequently depend on a huge number of uncertain parameters, but the amount of data per parameter is small. We propose a framework for optimization in this small-data/high dimensional regime. We combine ideas from empirical bayes and compound decision problems to 1) pool information across parameters and 2) exploit optimization structure. We prove our approach is never worse than, and often better than, “estimate-then- optimize,” sample average approximation, and conventional high-dimensional statistics techniques. As the number of uncertain parameters grows and the amount of data per parameter is fixed, our approach performs comparably to the best-in-class estimator. impact of market share and marginal cost are further discussed. 5 - A Bibliometric Analysis on Human Reliability Analysis Yangyang Chang, Tsinghua University, ZJ.17#, Tsinghua University, Haidian, Beijing, 100084, China, changyy15@mails.tsinghua.edu.cn, Pingke Li TE52

2 - Applications of the Wasserstein Metric in Robust Transportation Problem

John Gunnar Carlsson, University of Southern California, 3750 McClintock Avenue, Los Angeles, 90089, United States, jcarlsso@usc.edu Recent research on the robust and stochastic travelling salesman problem and the vehicle routing problem has seen many different approaches for describing the region of ambiguity, such as taking convex combinations of observed demand vectors or imposing constraints on the moments of the spatial demand distribution. One approach that has been used outside the transportation sector is the use of statistical metrics that describe a distance function between two probability distributions. 3 - Robust Multiclass Queuing Theory for Wait Time Estimation in Resource Allocation Systems Phebe T. Vayanos, University of Southern California, OHE 310L, University Park, USC, Viterbi School of Engineering, Los Angeles, CA, 90089, United States, phebe.vayanos@usc.edu, Chaithanya Bandi, Nikolaos Trichakis We present a data-driven optimization approach to estimate wait times for individual patients in the U.S. Kidney Allocation System, based on the very limited system information that they possess in practice. To deal with this information incompleteness, we develop a novel robust optimization analytical framework for wait time estimation in multiclass, multiserver queuing systems. We calibrate our model with highly detailed historical data and illustrate how it can be used to inform medical decision making and improve patient welfare. 361F Behavioral Operations Contributed Session Chair: Abhishek Sharma, Indian Institute of Management-Rohtak, Rohtak, India, aabhisheksharma01@gmail.com 1 - A Cooperative Game with Envy Ziteng Wang, Assistant Professor, Northern Illinois Universtiy, 590 Garden Road, EB 240, DeKalb, IL, 60115, United States, zwang3@niu.edu This research proposes an envy-incorporated cooperative game model and investigates the envy effects on players’ coalition-forming and payoff-allocating decisions. A player’s utility depends on her/his own payoff and the envy toward other players, inside and outside the same coalition, with higher payoffs. Envy core is defined to characterize stable coalition structures and payoff allocations of this new game. Conditions for the envy core to be nonempty are provided. The relative significance of in-coalition envy and out-coalition envy is shown to be a key factor to the form of the envy core. Application to a simple gameshows that envy may significantly change players’ decisions. 2 - Anchoring and Demand Chasing in Stockout Avoidance Decision Making Shivom Aggarwal, IE Business School, IE University, Instututo de Empresa S.L., CIF: B82334319, Calle de Maria de Molina, 12 Bajo, Madrid, 28006, Spain, saggarwal@faculty.ie.edu, Antti Tenhiala Even though, virtually every major supermarket chain uses an automatic store replenishment (ASR) system to optimize replenishment timing and order quantities; stock-outs (SO) still cost retailers billions of dollars every year in lost sales. In this study we focus on SO avoidance and argue that (i) augmentations of ASR proposals are driven by anchoring and demand chasing, and (ii) deviations triggered by these two biases are generally unsuccessful in avoiding stock-outs, leading to unnecessary inventory inflation. We test our hypotheses with proprietary data from a European high-end supermarket chain. Our study highlights the consideration of biases as critical and distinct from other drivers. 3 - A Summary and Comparison of Common Methods Variance Techniques Xiaotong Liu, University of North Texas, 2411 S Interstate 35 E, TE53

Apt 1218, Denton, TX, 75010, United States, xiaotong.liu@unt.edu, Victor R. Prybutok

In most empirical studies, the existence of common methods variance (CMV) is consideration for which the data should be tested. However, some challenges exist about which technique researchers should use to test CMV for data from a single instrument. This research uses examples to summarize and compare several tests related to CMV.

429