Informs Annual Meeting 2017

TA45

INFORMS Houston – 2017

5 - Statistical Techniques for Modeling and Simulation Validation Laura Freeman, IDA, 4850 Mark Center Drive, Alexandria, VA, 22311, United States, lfreeman@ida.org, Kelly Avery Modeling and simulation (M&S) can augment live testing to provide a more complete evaluation of system performance. It is essential that the usefulness and limitations of the M&S are well characterized, and uncertainty quantified. This talk highlights statistical techniques to compare M&S output with live test data. We will show how design for computer experiments can be used to efficiently cover the simulation domain and inform live testing. A simulation study shows that regression analysis is the most powerful method when paired with experimental design techniques. A Defense system case study shows highlight real world implications. 360D Revenue Management, Pricing Contributed Session Chair: Tulay Flamand, Colorado School of Mines, Golden, CO, United States, tflamand@mines.edu 1 - Tell Me What I Want: A Study of Personalized Assortment Planning for Learning Consumers Yulia Vorotyntseva, University of Texas at Dallas, 2200 Waterview Parkway, Apt 1627, Richardson, TX, 75080, United States, yxv120230@utdallas.edu, Canan Ulu, Dorothee Honhon We model the retailer’s and consumer’s simultaneous learning about the consumer’s idiosyncratic preferences in a given product category. In each period the retailer chooses the assortment of products to offer learns about the consumer’s preferences by observing her choice. The consumer picks at most one product, gets a noisy signal about its utility and updates her beliefs on her own preferences in Bayesian fashion. We derive structural properties of the firm’s optimal assortment policy and quantify the value of information about the consumer’s experience, such as feedback surveys. 2 - Novel Ways to Leverage Consumer Choices through Optimization James Farmer, President of The Harvest Group, 5500 West JB Hunt Drive, Suite 720, Rogers, AR, 72758, United States, jfarmer@harvestgroup.com Organizations struggle with how to determine appropriate spend. This dynamic reinforces the behavior of beginning with the “what did we do last year” approach - that reduces optimal results. By taking a different perspective that begins bottom-up and ends with practical engineering, brands and retailers can discover more growth. Data driven feedback loops provide executives with an ability to optimize past results based on predictive analytics. This dynamic re- allocation of resources guides managers and executives toward focusing investment where business can improve. 3 - Designing a Centralized Distribution System for Omnichannel Retailing Jia Guo, the University of Alabama, 311 Reed Street, Apt 8, Tuscaloosa, AL, 35401, United States, jguo23@crimson.ua.edu, Burcu B.Keskin Shifts in consumer spending toward services and experiences as well as the competition from discounters and online stores force the traditional retailers to reconsider their supply chain designs. We consider a two-stage stochastic integer programming model for designing an omnichannel network integrating brick- and-mortar stores with an online channel by filling online orders through shipping from nearby stores; by offering their customers the option to buy online and pick up in-store or curbside. Using a single period model, we evaluate the effectiveness of these cross-channel strategies for improving profitability, moving store inventory and avoiding markdowns. 4 - Simulation and Analysis of a Grocery Store Layout Jessica Peggy Dorismond, University at Buffalo, 17DA Creekside Village, Buffalo, NY, 14212, United States, jpdorism@buffalo.edu, Jose Walteros, Rajan Batta The purpose of this research is to utilize optimization and simulation modeling to yield an optimal supermarket layout. This study focuses on how to optimize the layout of a supermarket in order to increase its gross profit via the maximization of impulse sales. In most supermarkets many items often get unnoticed because on average customers only walk one-third of the store. Since customers use tangible products as a memory cues, increasing the visibility of certain items will prompt customers to purchase items that are not included on their shopping list. TA45

5 - Sales Assistance Search and Purchase Decisions an Analysis using Retail Video Data Aditya Jain, Baruch College, Zicklin School of Business, 55 Lexington Ave, Suite 9-240, New York, NY, 10010, United States, aditya.jain@baruch.cuny.edu, Nils Rudi, Sanjog Misra We investigate the roles of sales assistance and search in driving customer’s purchase decision using unique observational video data from retail stores. Our analysis reveals that both sales assistance and search play substantial roles which differ based on the context of specific decisions—search has a more dominant role in purchase incidence, whereas the latter in conditional expenditure. 6 - Retail Analytics on Store-wide Shelf Space Allocation Tulay Flamand, Colorado School of Mines, Engineering Hall 816 15th Street Office #313, Golden, CO, 80401, United States, tflamand@mines.edu, Ahmed Ghoniem, Bacel Maddah We investigate how to optimize the store-wide shelf space allocation of retailers in a way that guides in-store traffic and induces impulse buying. A predictive model for in-store traffic is formulated based on our data analysis and is embedded into an optimization model that prescribes shelf space allocation solutions. 360E Scheduling in Operations Management Invited: Project Management and Scheduling Invited Session Chair: Chelliah Sriskandarajah, Texas A&M University, College Station, TX, 77843-4217, United States, chelliah@mays.tamu.edu Co-Chair: Yunxia Zhu, Rider University, Pennington, NJ, 08534, United States, yuzhu@rider.edu 1 - Outpatient Appointment Block Scheduling under Patient Heterogeneity and Patient No-shows Yunxia Zhu, Rider University, 101 tuxford ct, Pennington, NJ, 08534, United States, yuzhu@rider.edu, Seung Jun Lee, Gregory R. Heim, Chelliah Sriskandarajah We study outpatient appointment block scheduling policies for single providers under conditions of patient heterogeneity in service times and patient no-shows. The objective is to find daily appointment schedules that minimize a weighted sum of patients’ waiting time, the physician’s idle time, and the physician’s overtime. We contribute by suggesting new effective sequential block scheduling procedures motivated by actual outpatient clinic practices across the globe and grounded in the successful Toyota Production System load smoothing approach. 2 - Lateral Transshipment Strategy with Customer Switching Ying Li, Texas A&M.University, College Station, TX, 77843, United States, yli@mays.tamu.edu, Wenjing Shen, Xinxin Hu, Yi Liao We consider inventory replenishment and transshipment decisions at two retail locations in a single-period setting while allowing the possibility of customers not staying with their initial retail location. Our research identifies the optimal replenishment and transshipment policy. In particular, we show that the surplus retailer should reserve part of its inventory for switching demand, as opposed to satisfy all transshipment demand. Through numerical studies, we examine the impact of random customer switching on profitability and demand fill rate, as well as the value of the optimal transshipment policy. 3 - Network Seeding on Social Media Rakesh Reddy Mallipeddi, Texas A&M.University, 320 Wehner - 4217, Mays Business School, Dept of Info&Operations, College Station, TX, 77843-4217, United States, rmallipeddi@mays.tamu.edu, Subodha Kumar, Yunxia Zhu, Chelliah Sriskandarajah With explosive growth in number of users, marketers have embraced social media as a preferred platform to diffuse and advertise their products. In this study, we propose analytical models to identify influencers (or network seeders) to maximize the reach and effect of an advertisement (or seed) on social media platforms. 4 - A Framework for Analyzing the U.S. Coin Supply Chain Yunxia Zhu, Rider University, 101 Tuxford Ct, Pennington, NJ, TA46 This study addresses operational issues within a Coin Supply Chain (CSC) and provides an optimal/near-optimal operating policy for the Federal Reserve System and Depository Institutions to increase their efficiency and cost-effectiveness in ordering, producing, packaging, distributing, and managing inventory of coins. We treat the U.S. CSC as a closed-loop supply chain from both supply and demand side perspectives. We devise and perform an extensive computational study to answer managerially relevant questions in the context of improving the efficiency of the U.S. CSC. 08534, United States, yuzhu@rider.edu, Yiwei Huang, Subodha Kumar, Chelliah Sriskandarajah, Bala Shetty

286