Informs Annual Meeting 2017

POSTER SESSION

INFORMS Houston – 2017

13 - Job Shop to Cellular Manufacturing: An Implementation in RF/microwave Manufacturing Okan Karatay, Production Manager, Aselsan, Konya Yolu 8 km Ogulbey Mh. 3051 sk no:3, Ankara, 06830, Turkey, okaratay@aselsan.com.tr, Yuce Cinar Baysal Nowadays, one of the effective ways of dealing with manufacturing drawbacks in a high mix low volume environment is the Cellular Manufacturing (CM). In this study, we first propose a manufacturing cell that is designed to replace the traditional job shop layout of Aselsan Inc. based on lean perspective. Secondly, we focus on the scheduling that creates the continuous flow and provides the line balance. In addition, employee originated challenges of adopting CM are also deeply investigated. Thirdly, we evaluate the performance of the CM with the conventional production layout by using the company’s KPI. Finally, we derive some suggestions on how human aspects of CM can be improved for future studies. 14 - Learning in Discrete Optimization Elias B.Khalil, PhD Student, Georgia Tech, North Avenue, Atlanta, GA, 30332, United States, lyes@gatech.edu The branch-and-bound algorithm for solving Mixed Integer Programs (MIP) involves decisions that are currently addressed heuristically. Instead, we propose to use machine learning (ML) to make better-informed, input-specific decisions in MIP branch-and-bound. For branching and primal heuristic selection, our ML approaches significantly improve solver performance on heterogeneous MIPLIB instances and certain homogeneous families of instances. For some classical combinatorial problems, we propose a deep graph embedding framework for learning powerful greedy heuristics, and show that the learned heuristics are competitive with existing heuristics for VertexCover, MaxCut and TSP. 15 - Development of Accelerated Reliability Demonstration Test Plan under a Nonlinear Random-coefficients Model Seongjoon Kim, Assistant Professor, Chosun University, 3F Dept. of IE, 2nd Engineering Building, 309 Pilmun-daero, Dong-gu, Gwangju, 61452, Korea, Republic of, seongjoon.kim@chosun.ac.kr, Suk Joo Bae Due to tight time-to-market constraints, manufacturers have to assure product reliability in a short period of time and with limited resources. In this circumstance, reliability demonstration test(RDT) is widely used to demonstrate whether a product meets a pre-determined reliability level. In this research, we propose a method for Reliability Demonstration by Accelerated Degradation Test in the context of a nonlinear random-coefficients model. The proposed method is illustrated with display back-light unit(BLU) case and is applied to demonstrate the luminosity reliability. 16 - Millennials’ Social Networking Behavior Anita Lee-Post, University of Kentucky, Department of Marketing and Supply Chain, 435Q.Business & Economics Bldg, Lexington, KY, 40506, United States, dsianita@uky.edu The purpose of this research is to further our understanding on Millennials’ social networking behavior. Past literature has characterized Millennials as narcissists with a high level of technological competency but a low level of privacy concerns and commitment. A correlation study is sought to investigate the relationships among the characteristics of Millennials and their social networking behavior using a survey methodology. Theoretical and practical implications of findings from this study will be presented. 17 - A Semiparametric Trajectory Model to Predict Cognitive Decline in Lewy Body Dementia Xiaoxia Li, North Dakota State University, Fargo, ND, 58102, United States, xiaoxia.li.lisa@gmail.com, Cannan Bilen-Green, Kambiz Farahmand, Linda Langley Lewy bodies dementia (LBD) is a neurocognitive disorder that result in memory loss and other cognitive or functional decline. LBD progression is highly diverse, and the symptoms vary differently from case to case. However, it is essential to map the cognitive decline to the severity of the impact of the symptoms on daily life. Clinical Dementia Rating SUM score (CDR SUM score) is a comprehensive evaluation tool. In this study, we built a semiparametric model showed robustness in both explaining the data and predictions. The study results revealed the factors associated with cognitive decline rate in terms of CDR SUM score, and gave implications on accurate CDR SUM score prediction. 18 - Handling Passenger Flows and Transfers in Flight-to-gate Reassignment Procedures During Schedule Disruptions Moschoula Pternea, PhD Candidate, University of Maryland, College Park, MD, 20742, United States, mpternea@umd.edu, Ali Haghani The problem of reassigning flights to airport gates arises when unexpected disruptions render the planned assignment infeasible or impractical. In this context, one of the biggest challenges is to handle passenger flows, which are involved in the prevention of missed transfers, as well as the optimization of total walking distance for connecting passengers. We present an IP-based framework to handle the computational complexity of passenger flows by considering the layout of the airport and scheduled flights of the aircraft for the same day in order to prevent delay propagation across the whole network. Our experiments reveal the tradeoffs between different aspects of the solution quality.

19 - Assessing the Health and Wellness Outcomes of Infants Born to Teen Mothers Toyya A.Pujol, PhD Student, Georgia Institute of Technology, Atlanta, GA, United States, pujol@gatech.edu, Nicoleta Serban, Julie L.Swann The research will explore the impact of teen pregnancy on the health and wellness outcomes of infants. Causal effects of teen pregnancy were found by matching teen and adult mothers and performing a pairwise comparison for each outcome. The mothers were matched using demographics, health status, and Medicaid eligibility. Outcomes of the infants were restricted to the first year of birth and included addiction rates, health status, foster care, number of Emergency department and wellness visits. Analysis of outcomes includes using the Poisson means test and binomial proportion test. 20 - An Evolutionary Approach to Finding Improved Airplane Boarding Order Ed Ramsden, Murata Power Solutions, 1080 County Street, Attleboro, MA, 02703, United States, ed@edscave.com This project proposes a method for improving seat-to-zone assignments for boarding passenger aircraft. A simulation model is used to generate overall boarding times for different passenger seat-to-zone assignments, and an evolutionary algorithm uses the results of the simulation model to search for seat- to-zone assignments which result in minimal boarding times. 21 - Intensity Based Resilience Metric: An Exploration in Social and Physical Systems Daniel Hernando Romero, University of South Florida, 4411 Shady Terrace Ln, Unit A, Apt 212, Tampa, FL, 33613, United States, danielromero@mail.usf.edu Daniel Hernando Romero, Universidad del Norte, Barranquilla, Colombia, danielromero@mail.usf.edu, Alex Savachkin A new resilience metric is proposed and evaluated in social and physical systems. The intensity based metric includes the disruption magnitude to enable a fair comparison when different events are studied. The analysis is carried out in a community under a pandemic influenza outbreak and in a supply chain. The George Saman, California State University-Fullerton, Fullerton, CA, United States, georgesaman@csu.fullerton.edu, Paulina Reina Empirical data at a freeway site was analyzed to investigate features of traffic state transitions from free-flow to congested state and vice versa. It was observed that transition durations range approximately from 15 to 35 minutes on average during congestion onset and clearance. Results also suggest that the rate of transition growth increases as the transition zone travels upstream of the bottleneck and that merges and diverges have no effect on the transition mechanism. These findings can have significant implications on the effectiveness of traffic management strategies during periods of transitions. 23 - An Approach Towards Automation of Model Selection Mohammadali Shirazi, Texas A&M University, College Station, TX, United States, alishirazi@tamu.edu, Dominique Lord We present an alternative to typical Model Selection strategies. Unlike typical methods, our approach looks at the characteristics of data to come with the best model. Our approach is easy to use and do not need any post-modeling inputs. It is particularly most suitable for instances when many alternatives with different complexities exist or the analyst deals with big data. 24 - Sustainable Supply Chain Management and the Issue of Social Sustainability Bruno Silvestre, Associate Professor, University of Manitoba, Winnipeg, MB, Canada, b.silvestre@umanitoba.ca, Dafne Moraes The social dimension of sustainability has been poorly investigated when compared to the environmental dimension in sustainable supply chain management studies. This lack of attention is problematic for the theory and practice of sustainable supply chains. We employ the concepts of supply chain engagement and motivation to shed light on the topic and improve our understanding of supply chain social sustainability. We selected eight cases of social initiatives undertaken by companies and used them as examples of the four different supply chain approaches to social sustainability. We finish the paper with implications for the practice and contributions to the supply chain management theory. results are contrasted with traditional average performance metrics. 22 - Spatio-temporal Characteristics of Freeway Traffic State Transitions

369