2015 Informs Annual Meeting

SB53

INFORMS Philadelphia – 2015

SB51 51-Room 106B, CC Retail Operations Sponsor: Manufacturing & Service Operations Management Sponsored Session Chair: Amy Pan, Assistant Professor, University of Florida, Dept. of ISOM, Warrington College of Business Administr, Gainesville, FL, 32608, United States of America, amy.pan@warrington.ufl.edu Co-Chair: Dorothee Honhon, Associate Professor, University of Texas at Dallas, 800 W. Campbell Road, Richardson, TX, 75080, United States of America, Dorothee.Honhon@utdallas.edu 1 - Do MOOCs Work in Retail? Improving Store Execution through Online Learning Santiago Gallino, Tuck School of Business, 100 Tuck Hall, Hanover, NH, United States of America, santiago.gallino@tuck.dartmouth.edu, Youran Fu, Serguei Netessine Conversion of retail store traffic into sales depends heavily on store execution. Although a critical piece, performance improvement of sales associates through training has received relatively scant research effort. We focus on a very specific type of training: online learning. Available to date analysis of MOOCs indicates poor engagement of the participants and questionable outcomes. Our analysis shows that online training has a strong positive impact on employee sales performance. 2 - Backroom Effect in Shelf Space Optimization Zumbul Atan, Assistant Professor, TU/e, Eindhoven, Eindhoven, Netherlands, Z.Atan@tue.nl, Nesim K. Erkip, Tim Huh Most retailers use backrooms as extra storage space. Backrooms are necessary when inventories do not fit to the available shelf space. We study a continuous review (r,q) policy with limited shelf capacity. We determine the optimal policy parameters and the optimal shelf space and quanitufy the benefits of considering the existence of backrooms when making replenishment decisions. 3 - Probabilistic Selling for Vertically Differentiated Products Quan Zheng, Student, University of Florida, Department of ISOM, Gainesville, FL, 32611, United States of America, quanzheng@ufl.edu, Janice E. Carrillo, Amy Pan We investigate the probabilistic selling strategy for vertically differentiated products. Both deterministic and stochastic demand models are explored. The results show that capacity constraints and demand uncertainty play important roles. The probabilistic products could be offered alone in the deterministic demand model while they are offered alone or with the low quality products in the stochastic demand model. 4 - The Impact of Consumer Search Cost on Assortment Planning and Pricing Ruxian Wang, Johns Hopkins University, 100 International Dr, Baltimore, MD, 21202, United States of America, ruxian.wang@jhu.edu, Ozge Sahin We incorporate search cost into consumer choice: a consumer first forms her consideration set; she evaluates and chooses the highest-utility product. We propose the k-quasi attractiveness-ordered assortment and show that it is arbitrarily near-optimal. Assortment problems are generally NP-hard, so we develop efficient exact and approximation algorithms. We propose a new pricing strategy and investigate its optimality: the quasi-same-price policy with a same price for all products except one. SB52 52-Room 107A, CC Uncertainty and Performance of Service Processes Sponsor: Service Science Sponsored Session Chair: Genady Grabarnik, St. John’s University, 8000 Utopia Parkway, Queens, NY, 11439, United States of America, grabarng@stjohns.edu 1 - Palm Khinchine Theorem and Performance Evaluation under Uncertainty Genady Grabarnik, St. John’s University, 8000 Utopia Parkway, Queens, NY, 11439, United States of America, grabarng@stjohns.edu, Yefim Haim Michlin, Larisa Shwartz We refine estimates for Pam Khinchine theorem, which allows us to tighten boundaries for comparison performance evaluation in sequential testing

2 - Recommending Resolutions for Monitoring Tickets in Automated Service Management Larisa Shwartz, lshwart@us.ibm.com, Genady Grabarnik We study the problem of automated resolution recommendation for monitoring tickets. We analyzed monitoring tickets from a production service infrastructure and identified a vast number of repeated resolutions. We improve the similarity measure in comparison to prior work by utilizing both the description and resolution information from historical tickets via a topic-level feature extraction using the LDA model. Furthermore effective similarity measure is learned using metric learning. 3 - Comparison Performance Evaluation for Services Systems Yefim Haim Michlin, Senior Research Fellow, Technion - Israel Institute of Technology, Faculty of IEM, Technion City, Haifa, 32000, Israel, yefim@technion.ac.il, Ofer Shaham, Genady Grabarnik Enterprises spend significant effort on introduction of innovations. We suggest a novel method based on Wald’s sequential test idea to identify performance gain by running comparison test. The advantage in comparison to the existing tests suggested in textbooks and standards are: more strict limitation on maximum number of trials, wider range of failure ratio, variable risks and wider discrimination ratio. Suggested methodology is oriented on usage by product designers and quality engineers. SB53 53-Room 107B, CC From Behavioral/Experimental Economics to Behavioral Operations: Opportunities and Challenges Sponsor: Behavioral Operations Management Sponsored Session Chair: Tony Haitao Cui, Associate Professor, University of Minnesota, Carlson School of Management, Minneapolis, MN, 55455, Teck-Hua Ho, William Halford Jr. Family Professor of Marketing, University of California, Berkeley, Berkeley, CA, 94720-1900, United States of America, hoteck@haas.berkeley.edu The notion of equilibrium is central to economic analysis. Standard equilibrium analysis relies on strong assumptions of human cognitive ability, which often do not match actual behavior. Based on this, we explain why management science should embrace non-equilibrium models. We describe recent research on them and discuss the promise that they hold based on their ability to predict better than equilibrium analysis, behavior observed in more than 100 laboratory experiments and field settings. 2 - Economics and Operations: Similarities and Differences Rachel Croson, Dean, College of Business, University of Texas at Arlington, Arlington, TX, 76019-0377, United States of America, croson@uta.edu This presentation will identify some of the similarities and differences between the underlying fields of economics and operations, with a goal of highlighting how experimental and behavioral research should be differentially applied to each field. 3 - The Effect of Bargaining on Testing Operations Management Models Elena Katok, Ashbel Smith Professor of Supply Chain Management, University of Texas at Dallas, 800 W. Campbell Rd., Dallas, TX, 75080, United States of America, ekatok@utdallas.edu OM models that involve strategic interactions are usually silent about the bargaining process. The outcome of laboratory tests of these models, however, can critically depend on how the bargaining is implemented. I will discuss the challenges and opportunities of incorporating bargaining in laboratory tests. United States of America, tcui@umn.edu 1 - Non-equilibrium Models in Games

83