2016 INFORMS Annual Meeting Program

MC94

INFORMS Nashville – 2016

MC94 5th Avenue Lobby-MCC Technology Tutorial: GAMS/LINDO Technology Tutorial 1 - GAMS: Introduction To Modeling In GAMS Steven P Dirkse, GAMS Development Corporation, Washington, DC, United States, sdirkse@gams.com We demonstrate many of the capabilities of the GAMS software as we start with a simple optimization model and build it out by adding nonlinear and integer variables to the model and connecting it with a GUI in a sample application. 2 - LINDO: Optimization Modeling Made Easy Mark A Wiley, LINDO Systems Inc, 1415 No Dayton Street, Chicago, IL, 60622, United States, mwiley@lindo.com, Gautier Laude Monday, 3:10PM - 4:00PM Davidson Ballroom-MCC Omega Rho – 40th Year Anniversary Panel Plenary Session Chair: Graham Rand, Lancaster University, United Kingdom, Lancaster, LA1 4YX 1 - Omega Rho - 40th Year Anniversary Panel Graham Rand, Lancaster University, Lancaster, United Kingdom, g.rand@lancaster.ac.uk After a brief introduction to Omega Rho, International Honor Society for Operations Research and Management Science, as it celebrates its 40th birthday, four of its distinguished lecturers will revisit their lectures. All four were in the first group of INFORMS Fellows, created in 2002 2 - Panelist John R. Birge, Jerry W. and Carol Lee Levin Professor of Operatio, University of Chicago, Booth School of Business, Chicago, IL, United States, John.Birge@ChicagoBooth.edu 3 - Panelist John D. Little, Massachusetts Institute of Technology, M.I.T. Sloan School Of Management, Room E62-534, Cambridge, MA, 02142, United States, jlittle@mit.edu 4 - Panelist Ralph Keeney, Duke University, San Francisco, CA, United States, keeneyr@aol.com 5 - Panelist Alfred Blumstein, Carnegie Mellon University, Heinz College - Hamburg Hall, Pittsburgh, PA, United States, ab0q@andrew.cmu.edu Monday Plenary

2 - State Transition Modeling For An Interdisciplinary Pain Management Program Nilabh Ohol, The University of Texas at Arlington, nilabh.ohol@mavs.uta.edu We discuss state transition modeling for an adaptive interdisciplinary pain management program at the University of Texas Southwestern Medical Center at Dallas. Challenges include data collection and preparation, endogeneity, and statistical modeling for optimization. Different modeling approaches will be presented, including linear and piecewise linear regression, piecewise linear networks, and regression splines models. 3 - Challenges In State Transition Modeling For A System Of Electric Vehicle Charging Stations Ying Chen, The University of Texas at Arlington, ying.chen@mavs.uta.edu In order to supervise the running of plug-in hybrid electric vehicle (PHEV) charging station intelligently, approximated dynamic programming (ADP) algorithm is proposed to control this system, which is equipped with a distributed energy storage system charged by solar power, wind power and electricity from the power grid. The sampling of state space and state transition model are the critical parts to build a converged future value function (FVF) in ADP considering the dimension of state space and multicollinearity issue between state variables. In PHEV charging station control problem, the objective is to minimize the operational cost. 4 - Multicollinearity In State Transition Modeling Victoria C. P. Chen, University of Texas, 701 S. Nedderman Drive, Arlington, TX, 76019, United States, vchen@uta.edu, Bancha Ariyajunya, Ying Chen, Seoung Bum Kim Multicollinearity is known to have a negative impact on statistical modeling, specifically with respect to variance inflation. A state transition modeling approach based on orthogonalization of the state space is presented. Results are shown for a ground-level ozone pollution stochastic dynamic program.

MD02 101B-MCC Panel: Funding Issues at NSF Invited: NSF Invited Session

Moderator: Sheldon H Jacobson, University of Illinois, 201 N. Goodwin Avenue (MC258), Urbana, IL, 61801, United States, shj@illinois.edu 1 - Funding Issues At NSF: Broader Impact Changes Panelist: Sheldon H Jacobson, University of Illinois, shj@illinois.edu Discuss outcomes of recent NSF-sponsored workshop on Broader Impact, and its impact on future funding decisions. 2 - Funding Opportunities At The National Science Foundation Panelist: Diwakar Gupta, University of Minnesota and National Science Foundation, guptad@umn.edu 3 - Broader Impact at NSF Panelist: Sheldon Jacobson, University of Illinois, shj@illinois.edu

Monday, 4:30PM - 6:00PM

MD03 101C-MCC Daniel H. Wagner Prize Competition III Award Session

MD01 101A-MCC Data Mining for State Transition Modeling Sponsored: Data Mining Sponsored Session

Chair: C. Allen Butler, Daniel H Wagner Associates, Inc., 2 Eaton Street, Hampton, VA, 23669, United States, Allen.Butler@va.wagner.com 1 - IBM Cognitive Technology Helps Aqualia Reduce Costs And Save Resources In Wastewater Treatment Alexander Zadorojniy, IBM Research, IBM Haifa Research Lab, Haifa, Israel, Zalex@il.ibm.com, Segev Wasserkrug, Sergey Zeltyn, Vladimir Lipets This work takes a deep dive into operational management optimization problems in wastewater treatment plants. We used a constrained Markov Decision Process as the key optimization framework. Our technology was tested in a one-year pilot at a plant in Lleida, Spain, operated by Aqualia, the world’s 3rd-largest water company. The results showed a dramatic 13.5 percent general reduction in the plant’s electricity consumption, a 14 percent reduction in the amount of chemicals needed to remove phosphorus from the water, and a 17 percent reduction in sludge production.

Chair: Victoria C. P. Chen, The University of Texas at Arlington, Dept. of Ind., Manuf., & Sys. Engr., Campus Box 19017, Arlington, TX, 76019, United States, vchen@uta.edu 1 - A High-dimensional State Transition Development Framework For Deicing Activities At Dallas-fort Worth International Airport Zirun Zhang, FedEx, zhang.zirun@gmail.com For high-dimensional and complex systems, state transitions can be empirically represented from data to enable system simulation or optimization. This paper presents a data-driven framework for state transition development in the context of deicing/anti-icing activities at Dallas-Fort Worth (D/FW) International Airport. From study of the framework, the D/FW deicing system is stochastic, finite horizon and discrete-time, non-stationary, and non-convex with mostly continuous state variables.

208