Informs Annual Meeting 2017

TD45

INFORMS Houston – 2017

3 - Benefits of Optimal Testing Schedules for Sequential Adaptive Clinical Trials Alba Rojas-Cordova, Assistant Professor, Southern Methodist University, Caruth Hall 339, Dallas, TX, 75205, United States, alba@smu.edu, Ebru Korular Bish Existing sequential analysis methodologies can be applied to sequential clinical trials with interim analyses to preserve the statistical significance and power levels of a fixed sample size trial. These methodologies, however, do not prescribe the number of patients that need to be recruited and analyzed every period, nor incorporate financial constraints, and do not allow for learning. We utilize a stochastic dynamic programming model with Bayesian updates to derive an optimal testing schedule and an optimal futility threshold, which can be combined with a wide range of sequential stopping methodologies that address stopping for benefit. 4 - Optimal Post-donation Nucleic Acid Testing of Donated Blood Considering Viral Load Growth Curves and Donor Characteristics Hadi El-Amine, George Mason University, 4310 Cotswolds Hill Ln, Fairfax, VA, 22030, United States, helamine@gmu.edu, Ebru Korular Bish, Douglas R. Bish Blood product safety, in terms of being free of transfusion-transmittable infections (TTIs), is crucial. Nucleic Acid Testing (NAT) technology enables earlier detection of infections but is more expensive, hence, most blood centers administer NAT to pools of blood samples from multiple donors. Since some donor characteristics are uncertain, we develop a chance-constrained model that determines the optimal NAT pool sizes for various TTIs, considering both non-universal (where first-time donors undergo more extensive screening), and universal (i.e., common testing for all donors) strategies, so as to minimize the TTI risk, while remaining within the testing budget with a high probability. 360C Operations Management/Marketing Interface Contributed Session Chair: Lauri Saarinen, University of Lausanne, Lausanne, Switzerland, lauri.saarinen@unil.ch 1 - Simulation of Inventory Systems with Unknown Input Models Canan Gunes Corlu, Assistant Professor, Boston University, 808 Commonwealth Avenue, Boston, MA, 02215, United States, canan@bu.edu, Alp Akcay Assuming the availability of limited demand data, we consider the stochastic simulation of an inventory system with unknown demand distributions. Building on a nonparametric Bayesian approach, we propose a simulation replication algorithm that estimates a service level for the inventory system without making any assumptions on the functional form of the demand models. We illustrate our approach in a single-product inventory simulation. 2 - Newsvendor Problem and Approximate Solutions under Price Dependent Intensity Ahasan Harun, University of North Texas, 316 Fry Street, Apt 203, Denton, TX, 76201, United States, MdAhasanUddin.Harun@unt.edu Approximate solutions have been proposed to the newsvendor problem when the demand is a compound Poisson and with price dependent intensity. The distribution of the selling time may not necessarily follow an exponential distribution as is typically assumed. The normal distribution provides a good empirical fit for fast moving items. However, in practice, only the first two moments may be known. Examining approximations and the optimal price for order quantity items under limited information about distributions may provide insights into new guidelines for modeling this complex problem. 3 - Skill-based Performance Model of Heterogeneous Multi-skilled Workers in Seru Production Systems Yin Gai, Dongbei University of Finance & Economics, 217 Jianshan Street, Shahekou District, DaLian, 116025, China, gaiyin@dufe.edu.cn Effective cross-training for multi-skilled workers is viewed as a critical factor when it comes to enhancing the performance of seru production system (SPS). Responding to this importance, we develop a skill-based analytical model of the performance of heterogeneous multi-skilled workers accounting for individual learning/forgetting curves in different types of serus. The proposed analytical model has been successfully applied to a practical case using a discrete event simulation model, and the result justifies incorporating skill-based modeling into cross-training of SPS results in an accurate analyses and informed decisions of actual and future workforce performance. TD44

4 - Operations Research in the Management of Linen in Indian Railways Balaraman Rajan, Assistant Professor, California State University East Bay, 25800 Carlos Bee Blvd, Hayward, CA, 94542, United States, balaraman.rajan@csueastbay.edu, Ravichandran Narasimhan Indian railways operates 12,000 trains a day ferrying 23 million passengers a day. About 9,000 couches use linen and 1.8 million pieces are washed everyday by Indian railways. The infrastructure includes 41 mechanized laundries and estimated expense of Rs. 36 per linen set. This paper traces the life cycle of a linen, outsourcing options, capacity and their location, inventory, and flow management. It reviews the existing practice and identifies modelling opportunities to improve operations. 5 - Retailer’s Efforts on Quality Assurance under Manufacturer Encroachment Mao Yuan, Huazhong University of Science and Technology, Luoyu Road 1037, Wuhan, 430074, China, myuan@hust.edu.cn, Wanjiang Deng, Shihua Ma We study a two-echelon supply chain where a manufacturer produces and distributes products through an independent retailer or direct channel. The production may yield defective units, and the retailer conducts an inspection to assure quality. We exam the impact of manufacture encroachment on retailer’s quality assurance level, and the results show that manufacturer encroachment will induce the retailer to enhance quality assurance level,and the manufacturer may decrease the wholesale price. We also find that the encroachment may lead to win-win outcomes and lose-lose outcomes for the manufacturer and the retailer. 6 - Using Simulation to Bridge the Theory-practice Gap Lauri Saarinen, University of Lausanne, Lausanne, Switzerland, lauri.saarinen@unil.ch, Kyle D.Cattani, Katariina Kemppainen, Suzanne de Treville Scheduling a production facility around a product portfolio that combines products varying in time sensitivity makes it possible to combine responsiveness with effective deployment of capacity. This volatility-portfolio theory emerges from quantitative finance: Typical managers lack the background to carry out the calculations so hesitate to implement the theory, which is nonlinear and counterintuitive. We describe use of simulation modeling in two companies to help managers get comfortable with implementation. Managers then suggested additions to the model that gave insight into boundary conditions of the theory. 360D Retail Management Contributed Session Chair: Sajjad Farahani, University of Wisconsin-Milwaukee, Milwaukee, WI, United States, farahani@uwm.edu 1 - Production Planning of a Remanufacturing System with Variable Quality Returns in Reverse Logistics Environment TD45 Sajjad Farahani, PhD Candidate, University of Wisconsin Milwaukee, WI, 53211, United States, farahani@uwm.edu, Wilkistar A.Otieno In this paper, proposed a general framework for a remanufacturing environment and a mathematical model to maximize the profit by optimally deciding the quantity of products to be remanufactured and number of products disassembled or disposed. 2 - Optimization of Warranty Period and Product Price for Remanufactured Products using Mathematical Method Amir Kordijazi, Teaching Graduate Assistant, PhD Candidate, University of Wisconsin-Milwaukee, 1133 E Pleasant St, Warranty is an important factor in building a good manufacturer—consumer relationship. Manufacturers pursue to minimize warranty costs and at a same time they try to offer a warranty which promises acceptable product quality and reliability. This study considers a remanufactured electrical product under a tiered warranty policy and try to presents an optimal warranty period from the perspective of a manufacturer to maximize the total expected profits, while ensuring sustained consumer relation. Real data from a local company with a global supply chain was used to provide a numerical example. Apt # 103, Milwaukee, WI, 53202, United States, kordija2@uwm.edu, Yuxi Liu, Wilkistar A.Otieno

391