2016 INFORMS Annual Meeting Program

WD35

INFORMS Nashville – 2016

4 - Discretionary Service Line Design With Heterogeneous Customers Cuihong Li, University of Connecticut, cuihong.li@uconn.edu, Laurens G Debo We study discretionary service line design facing heterogeneous customers. For discretionary services, the longer the service time, the higher is the quality of the service. In the presence of variability, longer service times also create more congestion. Hence, a service firm needs to trade off congestion costs with value creation. We find that self-selection of heterogeneous customer might lead to distortion of the high-quality service, contrary to the classic product line design result of Moorthy (1984). WD31 202C-MCC Services in the Sharing Economy Sponsored: Manufacturing & Service Oper Mgmt, Service Operations Sponsored Session Chair: Laurens Debo, Dartmouth College, Tuck School of Business, Hanover, NH, 03755, United States, Laurens.G.Debo@tuck.dartmouth.edu Co-Chair: Luyi Yang, University of Chicago, Booth School of Business, Chicago, IL, 60637, United States, luyi.yang@chicagobooth.edu 1 - When Is Capacity Trading Among Consumers A Win-win To Consumers And Service Providers? We study a setting where consumers can trade among themselves unused capacity they purchased from a service provider (e.g., excess data on a mobile data plan). We examine implications for service provider profits and consumer surplus. 2 - Allocating Capacity In Bikeshare Systems Daniel Freund, Cornell University, Ithaca, NY, 14850, United States, df365@cornell.edu, Shane Henderson, David B Shmoys A Bikeshare system (BSS) allows users to rent and return a bike at any station within the system. The amount of usage data BSSs collect has increased greatly in the last decade, allowing us to develop data-driven methods to support their operations. In this talk we extend a continuous-time Markov chain model to allocate docks within a BSS so as to minimize the expected number of out-of- stock (OOS) events. We compute that quantity & efficiently find the allocations of bikes and docks that minimize it both over a finite horizon & at steady-state. Our work is used by NYC Bikeshare to redistribute bikes & (re-)allocate docks. 3 - Skill Screening In Large-scale Service Marketplaces Eren Basar Cil, University of Oregon, erencil@uoregon.edu, Gad Allon, Achal Bassamboo We consider a large-scale service marketplace where the moderating firm can run two skills tests on agents to assess if their skills are above certain thresholds. Our main objective is to evaluate the effectiveness of skill screening as a revenue maximization tool. We find that skill screening leads to negligible revenue improvements in marketplaces where agent skills are highly compatible. As the compatibility of agent skills weakens, we show that the firm starts to experience as much as 25% improvement in revenue from skill screening. Apparently, the firm can reap the most of these substantial benefits when it runs only one test. Behrooz Pourghannad, University of Minnesota, Minneapolis, MN, United States, behrooz@umn.edu, Saif Benjaafar, Jian-Ya Ding

associated with the stratification of state-space probabilities resulting in convex integer optimization models with entropy employed as either the objective to minimize or as a constraint with a maximum return objective. 2 - Quantifying The Contribution Of Seawalls To Mitigating Tsunami Damage Tom M Logan, PhD Pre-Candidate, University of Michigan, 1205 Beal Avenue, Ann Arbor, MI, 48109, United States, tomlogan@umich.edu, Jeremy D Bricker, Seth Guikema Seawalls are commonly used to defend against tsunamis, making it is essential we understand whether they truly mitigate damage. The north-east of Japan has been stuck by four tsunamis in the past 110 years. A model combining a cellular automaton and hydrodynamic models simulates how land development hypothetically changes with time and under different seawall height options. The insights will indicate which scenarios they provide physical protections and which scenarios require alternative action. 3 - Sample Sut Of Sample Inference Based On Wasserstein Distance Yang Kang, PhD Candidate, Columbia University, 1255th Amsterdam Ave SSW, RM901, New York, NY, 10027, United States, yangkang@stst.columbia.edu, Jose Blanchet We present a novel inference approach which we call Sample Out-of-Sample (or SOS) inference. Our motivation is to propose a method which is well suited for data-driven stress testing, in which emphasis is placed on measuring the impact of (plausible) out-of-sample scenarios on a given performance measure of interest (such as a financial loss). The methodology is inspired by Empirical Likelihood (EL), but we optimize the empirical Wasserstein distance (instead of the empirical likelihood) induced by observations. From a methodological standpoint, our analysis of the asymptotic behavior of the induced Wasserstein-distance profile David S Kim, Professor, Oregon State University, 204 Rogers Hall, School of Mech., Industrial and Mfg Eng., Corvallis, OR, 97331- 2407, United States, david.kim@oregonstate.edu, Xinyu Luo This research examines the current state-of-the-art in gauge capability analysis for destructive testing. Results are then presented that extend the specific destructive testing situations where gauge repeatability can be estimated. WD35 205A-MCC Retail Analytics & Optimization Sponsored: Manufacturing & Service Oper Mgmt, Service Operations Sponsored Session Chair: Tulay Flamand, University of Massachusetts Amherst, Isenberg School of Management, Amherst, MA, 01003, United States, varol@som.umass.edu 1 - Store-wide Shelf Space Analytics To Optimize Impulse Buying Tulay Flamand, University of Massachusetts Amherst, Amherst, MA, United States, varol@som.umass.edu, Ahmed Ghoniem, Bacel Maddah We address a store-wide retail shelf space allocation problem with the objective of promoting impulse buying. Basket data analysis is conducted using real data in order to calibrate a predictive model of in-store traffic. This predictive model is then embedded in a non-linear mixed-integer model in order to prescribe shelf space solutions that maximize impulse buying. 2 - Pricing And Inventory Decisions Of an Assortment Under Equal Profit Margins Bacel Maddah, American University of Beirut, Bliss Street, Beirut, Lebanon, bacel.maddah@aub.edu.lb, Hussein Tarhini, Melanie Jabbour We consider the interdependent decisions on inventory and pricing of substitutable products in an assortment. Within a newsvendor-type supply setting, we analyze the joint pricing and inventory decision problem of the retailer under the assumption that all products have equal profit margins. We derive several concavity and monotinicity results under two common consumer choice models, the logit and the nested-logit. We also present an extensive numerical study testifying to the near-optimlity of equal-margin pricing. function shows dramatic qualitative differences relative to EL. 4 - Destructive Testing Gauge Capability Analysis

WD32 203A-MCC Risk Analysis Contributed Session

Chair: David S Kim, Professor, Oregon State University, 204 Rogers Hall, School of Mech., Industrial and Mfg Eng., Corvallis, OR, 97331- 2407, United States, david.kim@oregonstate.edu 1 - Uncertainty, Entropy, And Ambiguity As Risk Measures In Decision Modeling For Portfolio Optimization David F. Rogers, University of Cincinnati, Department of Operations, Business Analytics, and Information Systems, Carl H. Lindner College of Business, 2925 Campus Green Drive, Cincinnati, OH, 45221-0130, United States, David.Rogers@UC.edu, George G Polak For a portfolio optimization setting, the risk from uncertain outcomes is typically considered. Other risk measures, including the choice of how to best employ the state-space probabilities and how to consider alternative probability functions are also important. Information entropy is incorporated for measuring the risk

465