2016 INFORMS Annual Meeting Program

MC69

INFORMS Nashville – 2016

3 - Analysis on Energy Efficient Switching Of Machine Tool With Stochastic Arrivals And Buffer Information Andrea Matta, Shanghai Jiaotong University, matta@sjtu.edu.cn Energy saving in production plants is becoming more and more relevant due to the pressure from governments to contain the environmental impact of manufacturing, and from companies to reduce costs. One of the measures for saving energy is the implementation of control strategies that reduce energy consumption during the machine idle periods. This talk will deal with switching policies that turn the machine off when production is not critical, and on when the part flow has to be resumed. A general policy is formalized by modelling explicitly the energy consumed at each machine state. MC67 Mockingbird 3- Omni Panel Discussion on Publishing in Quality and Reliability: The Editors’ Perspective Panel Session Moderator: Kaibo Wang, Tsinghua University, Beijing, China, kbwang@tsinghua.edu.cn 1 - Panel Discussion on Publishing In Quality And Reliability: The Editors’ Perspective Kaibo Wang, Tsinghua University, kbwang@tsinghua.edu.cn This panel brings journal editors to share their perspectives and experiences with the audience and answer questions pertaining to publication in Quality, Reliability and Data Sciences. Panelists are: Dr. Jianjun Shi, IIE Transactions; Dr. Fugee Tsung, Journal of Quality Technology; Dr. Peihua Qiu, Technometrics; Dr. Murat Caner Testik, Quality Engineering; Dr. Jing Li, Quality Technology and Quantitative Management. 2 - Panelist: IIE Transactions Jianjun Shi, Georgia Institute of Technology, jianjun.shi@isye.gatech.edu 3 - Panelist: QTQM Jing Li, Arizona State University, jing.li.8@asu.edu 4 - Panelist: Quality Engineering Murat Caner Testik, Hacettepe University, mtestik@hacettepe.edu.tr 5 - Panelist: Technometrics Peihua Qiu, University of Florida, pqiu@phhp.ufl.edu 6 - Panelist: Journal Of Quality Technology Fugee Tsung, HKUST, season@ust.hk MC68 Mockingbird 4- Omni Reliability Evaluation and Optimization from Complex Systems I Sponsored: Quality, Statistics and Reliability Sponsored Session Chair: Eunshin Byon, University of Michigan, College Station, MI, United States, ebyon@umich.edu Co-Chair: Qingyu Yang, Wayne State University, Detroit, MI, United States, qyang@wayne.edu 1 - A Space-time Autoregressive Model For Radar Images Under A Lagrangian Integration Scheme Xiao Liu, IBM T.J. Watson Research Center, Yorktown Heights, NY, United States, liuxiaodnn_1@hotmail.com This paper is concerned with the spatio-temporal modeling of two dimensional radar echo fields from a sequence of radar images. The method is useful for many environment- and energy-related problems. For example, the precipitation forecast, and the prediction of solar power production. 2 - Reliability Modeling For Continuous-state Systems Xinying Wu, Ohio University, wuxinying2009@gmail.com, Tao Yuan This talk presents a Bayesian hierarchical modeling framework for modeling the reliability and degradation of continuous-state systems composed of continuous- state components. Degradation modeling, degradation data analysis, system reliability prediction, and component important measures will be discussed.

3 - On The Probabilistic Site Selection Problem Yiwen Xu, North Dakota State University, Fargo, ND, 58102, United States, yiwen.xu6@gmail.com, Haitao Liao In this research, we study a site-selection problem in probabilistic networks where both nodes and edges are prone to be failed. To enhance the probability of connectivity from one node to another, options for adding multiple edges (i.e., edge-level redundancy) are considered. We formulate the mathematical programming problem and develop a method to solve the problem. Numerical examples are provided to demonstrate the problem and the use of the proposed solution methodology. 4 - A Physical-statistical Hybrid Model For Li-ion Battery Prognosis Nan Chen, National University of Singapore, isecn@nus.edu.sg The traditional PHM approaches for Li-Ion batteries relied on the experimental data, like battery capacity or impedance. We proposed a physical-statistical model to take full use of operational data, which are readily available, to model and predict the performance and reliability of Li-Ion batteries. Both numerical and case studies are constructed to demonstrate the effectiveness and promising futures of this physical-statistical model in the real applications. MC69 Old Hickory- Omni Military Resource Management Sponsored: Military Applications Sponsored Session Chair: Brian J Lunday, Air Force Institute of Technology, P.O. WPAFB, OH, 1, United States, brian.lunday@afit.edu 1 - Discrete Event Simulation-based Analysis Of Personnel Evaluation Policy The United States Army uses a forced ranking appraisal system, a practice largely abandoned in the private sector, in evaluating its officer corps. The psychological aspects of forced ranking evaluation systems have been well documented, but this study examines the mathematical aspects of how these systems can lead to misidentification of high-performing individuals. We show how the binomial distribution can explain many of the challenges, analyze human behavior in such a system, and create a discrete event simulation to analyze the effects of policy- driven constraints. 2 - Modeling And Forecasting Army Enlistments With Geographic Data Weighting, Principal Components Analysis, And Linear Regression Joshua McDonald, U.S. Army, Aberdeen Proving Ground, MD, United States, joshua.l.mcdonald10.mil@mail.mil Using ordinary least squares regression applied to geographically weighted panel data we forecast the production of Regular U.S. Army enlistments in 38 recruiting markets. We find that a set of five continuous independent variables obtained through principal components analysis plus categorical variables for markets and quarters of the fiscal year achieves effective 15-month forecasts; when forecasting independent variables, the models explain between 63% and 73% of the variation between actual and predicted data at the highest level of aggregation, depending on enlistment contract type. 3 - Optimal Design Of Piezoelectric Materials For Maximal Energy Harvesting Russell Nelson, United States Military Academy, West Point, NY, 10996, United States, russell.nelson@usma.edu, Hong Zhou, Susan Sanchez The DoD seeks alternative methods to produce electricity, thus decreasing dependence on fossil fuels and increasing combat power. Piezoelectric generators can produce alternative electrical power in isolated and austere conditions. We use three and six variable mathematical models to analyze piezoelectric generator power capabilities. Using mk factorial sampling, nearly orthogonal and balanced Latin hypercube (NOBLH) design, and NOBLH iterative methods, we find solutions to maximize piezoelectric generator power output. We further analyze our optimal results using robustness analysis techniques. Our results provide Lee A. Evans, University of Louisville, Louisville, KY, United States, laevan04@louisville.edu, Prajwal Khadgi

optimal material parameter and environmental designs. 4 - Risk Assessment In Robust Goal Programming Robert Hanks, Air Force Institute of Technology, robert.hanks@afit.edu

We investigate interval-based and norm-based uncertainty sets using cardinality- constrained robustness in the Robust Goal Programming (RGP) construct in addition to strict robustness using ellipsoidal uncertainty sets. Then, using utility theory, a decision maker’s (DM) view of risk is quantified via a utility function, which will be mapped back to relevant parameters of the varying uncertainty sets to model the DM’s risk attitude toward a robust solution. The findings offer theoretical contributions to the RGP framework and will be applied in a future endeavor to setting shipping rates for the United States Transportation Command’s customers as it pertains to revenue management.

205