2016 INFORMS Annual Meeting Program

MB79

INFORMS Nashville – 2016

5 - Procurement Strategy In Decentralized Supply Chains Under Supply Disruptions And Uncertain Demand Jie Tan, Huazhong University of Science and Technology, Wuhan, China, jietan@hust.edu.cn, Haijun Wang This paper considers a decentralized supply chain consisting of one retailer and one unreliable supplier under single-period setting. The retailer (buyer) has a chance of complete supply disruption and faces uncertain demand simultaneously. She (buyer) offers a contract (q, w), and upon it the supplier then chooses the optimal reliability level and pays for the corresponding cost of technology investment, to maximize his expected profit. Also the interaction between supplier’s reliability and production cost has been considered. We finally conduct a numerical analysis with the purpose of providing several managerial suggestions for order size and purchase price as well as reliability level. MB79 Legends G- Omni Health Care, Modeling VI Contributed Session Chair: David J Robb, Professor of Operations and Supply Chain Managemen, University of Auckland, Graduate School of Management, The University of Auckland Business School, Auckland, 1142, New Zealand, d.robb@auckland.ac.nz 1 - A Model For Diabetes Clinical Pathway Scheduling Hossein Badri, PhD Student, Wayne State University, 4185 4th Street, Manufacturing Building, Detroit, MI, 48202, United States, fq2529@wayne.edu, Kai Yang Clinical Pathway is a very efficient method to improve patient care, to manage clinical risks, and to improve resource efficiency. In this research we develop a model for the scheduling of the diabetes clinical pathway. The proposed model is tested using a real data set and the performance of the proposed model is analyzed. 2 - Modeling The Screening Of Post-traumatic Stress Disorder: A Criticism Of The Current Dichotomous Categorization Of Patients Navid Ghaffarzadegan, Virginia Tech, 1145 Perry Street, 231 Durham Hall, Blacksburg, VA, 24061, United States, navidg@vt.edu, Mohammad S. Jalali, Alireza Ebrahimvandi, Richard C Larson Post-traumatic stress disorder (PTSD) stands out as a major mental illness and screening policies have played a critical role in mitigating the effects of PTSD; however, little agreement exists on the optimal cutoff value—a threshold above which screening scores should be considered as PTSD positive. We develop a simulation model of PTSD screening which includes social forces that inhibit accurate screening. Our analysis shows that the results of long-term optimal policies are different than their myopic short-term results. With the presence of public stigma, current dichotomous screening policies are not much effective to minimize the number of false positive and false negative. 3 - Robust Multi Product Network Design For Blood Chain A Priority Queuing System Mohammad Abdollahi, PhD Candidate, Wayne State University, 4815 Fourth Street, Room 1067, Detroit, MI, 48202, United States, fq6861@wayne.edu, Kai Yang This study tries to investigate best network configuration in constructing a smooth and cost effective network considering all the modification corresponding to blood supply. Some highly important blood specific considerations such as blood type priority and lateral transshipment and perishability is considered in the proposed model. To cope with the existent uncertainty related to the blood processing and traveling times, a prioritized M/M/C and M/M/C models are considered for laboratories and blood banks, respectively. Moreover, an interval robust counterpart is proposed to deal with the uncertainty in the model corresponding to demand and cost and etc. 5 - Association Between Accident Rates And Daylight Saving Time Transitions

rates impacted in the week prior to the start and end of DST, i.e.,anticipation effects. Our results have implications for implementation and policy.

MB86 GIbson Board Room-Omni Marketing II Contributed Session 1 - Why Do Web Retailers And P&G Use “Postage Stamp” Pricing? First, this talk explains why some web retailers/mail order firms use zero or postage stamp type delivery charges but others set distance charges. A game theoretic model demonstrates the equilibrium decisions of a web retailer competing against reactive local retailers. I can show why Lands’ End and web furniture stores set flat prices but other specialty retailers charge by distance. Second, unrecognized in the literature is that P&G and other consumer products firms set location independent delivery charges to retailers. I present an economic model explaining this pattern. 2 - Marketing Strategies And Empirical Comparisons Between Agricultural Producer Community Structures Brian A. Bourquard, Purdue University, 403 W State Street, West Lafayette, IN, 47907, United States, bbourqua@purdue.edu, Allan W. Gray We study agricultural community structures and how influencers relate to information channel preferences. We use 1,300 agricultural producers’ ratings of information-source preferences and outside influencers from a nationally delivered survey to determine community structures with sociodemographic characteristics. We develop an empirical comparison of the producer communities’ purchasing and management behaviors. We also develop a methodology to predict the most effective marketing channels for input suppliers that accounts for the changing characteristics of agricultural producers. 3 - Modeling The Exposure Effect Of Customized Promotions On Shopping Frequency: A Quasi-experimental Analysis Russell Zaretzki, Associate Professor, University of Tennessee, 246 Stokely Management Center, Knoxville, TN, 37996, United States, rzaretzk@utk.edu, Amil Williamson, Mary Leitnaker Modern customer relationship management frequently involves the use of customized direct-to-consumer promotions. How well do such promotions work? We focus on the impact promotions have on the frequency of visits and analyze an unbalanced quasi-experimental crossover design that provides several varieties of customized and non-customized coupons to shoppers. Differences in trip frequency among consumers during periods with and without coupons are modeled. Carryover effects of promotion, decay rates of these effects, and positive feedback in the coupon allocation mechanism are also analyzed. MB94 5th Avenue Lobby-MCC Technology Tutorial: Optimization Direct/IBM Analytics Technology Tutorial 1 - Optimization Direct: Recent Advancements In Linear And Mixed Programing Give us The Capability To Solve Larger Optimization Problems Dr. Robert Ashford, Optimization Direct, Inc., 202 Parkway, Harrington Park, NJ, 07640, United States, rwa@optimizationdirect.com, Alkis Vazacopoulos CPLEX Optimization Studio solves large-scale optimization problems and enables better business decisions and resulting financial benefits in areas such as supply chain management, operations, healthcare, retail, transportation, logistics and asset management. In this tutorial using CPLEX Optimization Studio we will discuss modeling practices, case studies and demonstrate good practices for solving Hard Optimization Problems. We will also discuss recent CPLEX performance improvements and recently added features. 2 - IBM Analytics - 11:45am-12:30pm Xavier Nodet, IBM Decision Optimization, xavier.nodet@fr.ibm.com In this tutorial, you will learn about the new CPLEX and CP Optimizer engine features in the upcoming IBM CPLEX Optimization Studio release, including how to use the new features, how they can help you during the development of your model, and how they can speed up the resolution of your models. Phillip J Lederer, Professor, University of Rochester, Simon School of Bus. Admin., Rochester, NY, 14627, United States, Lederer@simon.rochester.edu

David J Robb, Professor of Operations and Supply Chain Managemen, University of Auckland, Graduate School of Management, The University of Auckland Business School, Auckland, 1142, New Zealand, d.robb@auckland.ac.nz, Thomas P Barnes

A third of nations have adopted Daylight Saving Time, presumably hoping that benefits exceed costs. Studies have found changes in accident rates associated with DST. Using data from 21 million New Zealand accidents during 2003-2015, we model accident rates as a function of various date-based predictors including days before/after the start and end of DST, day of week, and month of year. This is the first study to consider multiple accident categories (work, home, falls, sport, & road), and the first in the southern hemisphere. We find evidence of accident

174