Informs Annual Meeting 2017

TD16

INFORMS Houston – 2017

2 - Scheduling in Operating Theatres with an Entropic Index Jin Qi, Hong Kong University of Science and Technology, Room 5559C, Department of Industrial, Engineering and Logistics Management, Hong Kong, Hong Kong, jinqi@ust.hk, Han Ye In this talk, we propose a simple index to study the scheduling problem in the Operating Theatres (OT). We first use empirical data to explore the features of the scheduling issue in OT. Specifically, many surgeries can start earlier than the scheduled time. Then we formulate the scheduling problem such that the waiting time for each surgery and overtime should meet the service requirement. We propose an index named Entropic Deviation to capture both the variance and skewness of the uncertain surgery time. By simply ranking the Entropic Deviation for each type of surgery, we can make the sequencing decisions and achieve relatively good performance. 3 - Off-placing Patients in Hospital Wards: the Value of Flexibility in Healthcare Vahid Sarhangian, Univeristy of Toronto, Toronto, ON, Canada, sarhangian@mie.utoronto.ca, Carri Chan, Yuan Zhong When congestion is high, a common practice used in many hospitals is the off- placement of admitted emergency department (ED) patients to a non-primary inpatient ward (IW). We evaluate and compare the performance of practical flexible designs for routing patients from the ED to IW. We consider a multiclass queueing system with multiple server pools (wards) and allow the customers to be served by servers from a non-primary pool (determined through the flexible design) with a reduced service rate to capture the effect of off-placement. 4 - Inpatient Overflow: An Approximate Dynamic Programming Approach Pengyi Shi, Purdue University, 403 W. State St, Krannert School of Management, Kran 472, West Lafayette, IN, 47907, United States, shi178@purdue.edu, Jim Dai Inpatient beds are usually grouped into several wards, with each dedicated to a “primary” medical specialty. However, when a patient waits too long to get a primary bed, hospital managers may assign her to a non-primary bed though it is undesirable. Deciding when to use such “overflow” practice in real time can be challenging. We develop an approximate dynamic programming (ADP) algorithm to aid the decision making. A key step is to choose appropriate basis functions, for which we use a novel combination of fluid control and single-pool approximation. We demonstrate, via experiments in realistic hospital settings, that our ADP algorithm is remarkably effective in finding good overflow policies. 332E Retail Supply Chain Analytics Sponsored: Manufacturing & Service Oper Mgmt, Supply Chain Sponsored Session Chair: Joline Uichanco, University of Michigan, Ross School of Business, Ann Arbor, MI, 48109-1234, United States, jolineu@umich.edu 1 - The Role of Consumer Return Policies in Omnichannel Operations Leela Aarthy Nageswaran, Carnegie Mellon University, 308 GSIA, Tepper School of Business, 5000 Forbes Avenue, Pittsburgh, PA, 15213, United States, lnageswa@andrew.cmu.edu, Soo-Haeng Cho, Alan Scheller-Wolf We study the pricing and return policy decisions of an omnichannel retailer serving customers who differ in how they realize their valuation for the product — by inspecting in-store before purchase or by purchasing online and possibly returning misfit products. When the salvage value for returned products is low, we find that the firm may prefer full refunds to no refunds, in contrast to prior research showing that full refunds are never preferred at low salvage values in an online-only market. Furthermore, the optimal refund is often larger than the salvage value, which could explain generous refunds seen in practice. 2 - Optimizing SKU Selection for Promotional Display Space at Grocery Retailers Olga Perdikaki, University of South Carolina, Darla Moore School of Business, 1014 Greene Street, Columbia, SC, 29208, United States, Olga.Perdikaki@moore.sc.edu, Olga Pak, Mark Ferguson, Su-Ming Wu Promotional displays are a powerful tool to boost consumer engagement. We propose a multi-step methodology to identify a profit-maximizing selection of products for promotional display using grocery store sales transaction data. Our methodology results in a 5% to 23% improvement in profit when compared to an industry benchmark. 3 - Optimizing E-commerce Order Sourcing: A Big Data Analytics and Total Margin Optimization Solution Ali Koc, IBM.Research, IBM.Research, Yorktown Heights, NY, 10598, United States, akoc@us.ibm.com, Ajay Deshpande, Brian Quanz TD15

As online demand is growing rapidly, retailers are increasingly using their ship from store network with thousands of diverse nodes for online order fulfillment with conflicting business objectives. To overcome these challenges, we have developed a cloud-based solution for order fulfillment utilizing predictive analytics and multi-objective optimization to determine optimal total-margin fulfillment decisions. In this talk, we will present some key aspects of this solution and results. 4 - The Two-Phase Newsvendor Problem with Optimally-Timed Additional Replenishment Yale Herer, Israel Institute of Technology, Haifa, 32000, Israel, yale@technion.ac.il, Dina Smirnov, Assaf Avrahami We utilize in-cycle demand information, available from modern information systems, to optimize a newsvendor system with an option of an additional inventory review (AR) and replenishment at an endogenously determined, a priori set time during the period. We jointly optimize the AR timing and order quantities at each ordering opportunity. 332F Operations Management Contributed Session Chair: Xiaojun (Gene) Shan, University of Houston-Clear Lake, Houston, TX, United States, shan@uhcl.edu 1 - Total Cost Effects of the Interaction of Capacity, Periodic Release and Transportation for a Repair Shop Henny van Ooijen, Dr., Eindhoven University of Technology, P.O. Box 513, Eindhoven, 5600MB, Netherlands, h.p.g.v.ooijen@tue.nl A repair shop serves industrial customers that operate a commoditized capital good. The shop provides for availability of the capital good. Upon failure of a good, a substitute good, rented from a third party, is installed, and a logistics service provider is informed for transporting good. After arrival at the repair shop the failed good is repaired and returned to the customer; then the rent of the substitute good ends. We are interested under what condition periodic order release at the repair shop leads to cost benefits. 2 - Lean Six Sigma Techniques and a Practise in the Food Company Keeping the customer satisfaction in a high level and surviving with enough benefit level is getting harder to have at the same time with each passing day. Trading enterprises must minimize their costs while maintaining their certain quality level for completing this task. In this study, firstly Six Sigma and Lean terms have been examined and then integration of lean to six sigma studies has been examined and mentioned. In final part, non-quality reasons in a food company have been eliminated using Six Sigma techniques and optimum production rates have been designated via mathematical model created in Lean logic have been mentioned. 3 - Using Multi-variate Regression and Simulation to Identify a General Formula for Throughput of Flow Manufacturing Lines with Similar Stations Xiaojun (Gene) Shan, University of Houston-Clear Lake, Houston, TX, 77058, United States, shan@uhcl.edu, Samrawi Berhanu Accurately estimating throughput of a manufacturing line is important for planning. This paper use simulation and multivariate regression to estimate a generic equation to determine the throughput. Minitab was used to create random data that follow normal distributions: Station Speed, Mean time between failure, Mean time to repair, buffer capacity and line length. One thousand different random manufacturing lines were developed and modeled using PlantSim. The outputs were then used to run a multi-variate regression analysis to find an equation that best fits the data. A regression equation with R2 equal 92% was found. The station speed has the largest impact on throughput. 4 - Outsourcing Maintenance in the Presence of Cyber Threats Anh Ta, University of North Texas, Denton, TX, 76201, United States, Anh.Ta@unt.edu, Hakan Tarakci, Shailesh S. Kulkarni Outsourcing minimal repair and preventive maintenance is very important to any system. This study develops a model to maximize the profit of a manufacturer who utilizes an external contractor to maintain its manufacturing system. Later, we extend the study to a system with the existence of cyber threats. TD16 Harun Karakoç, Industrial Engineer, Erciyes University, Erciyes Teknopark Tekno-3 Binası No: 28, Kayseri, Turkey, harunkarakoc@gmail.com

381