2016 INFORMS Annual Meeting Program

TD01

INFORMS Nashville – 2016

Tuesday, 3:10PM - 4:00PM

Tuesday, 4:30PM - 6:00PM

Keynote Tuesday

TD01 101A-MCC New Methods in Data Mining Sponsored: Data Mining Sponsored Session

Davidson Ballroom A-MCC Optimizing the Future – Supply Chain at Amazon Keynote Session Chair: Anne Robinson, Verizon Wireless, Basking Ridge, NJ, anne.robinson@verizonwireless.com 1 - Optimizing The Future - Supply Chain at Amazon Jason Murray, Amazon, 410 Terry Avenue North, Seattle, WA, 89109, United States, n.a@na.org The retail supply chain of the future will be built on massive data, advanced analytics and innovative technology. At Amazon, we are constantly pushing the frontier in each of these areas to help create that future. Our vision is to move products at an unprecedented scale through the most technologically advanced supply chain possible, where intelligent optimization algorithms drive efficiency. To achieve this vision, we focus on three core pillars: research, technology, and business ownership. We develop new research, implement it in technology, and own the top- and bottom-line of the business. To be successful, we believe business ownership, research, and development must be tightly coupled. During this presentation, I will discuss our vision for the future of retail supply chain — where we have been, where we are, and where we plan to go. I will share some cases of research innovation and its integration with technology and business. These include inter-disciplinary modeling and optimization (from machine learning, statistics, simulation and optimization) to make Amazon’s supply chain more efficient. Finally, I will provide examples of some challenges we will need to overcome to make our vision a reality. Davidson Ballroom B-MCC Wagner Prize Winner Reprise Invited: Plenary, Keynote Invited Session Chair: Allen Butler , Daniel H Wagner Associates, Inc., Hampton, VA, allen.butler@va.wagner.com 1 - Wagner Prize Winner Reprise C. Allen Butler, Daniel H Wagner Associates, Inc., 2 Eaton Street, Hampton, VA, 23669, United States, Allen.Butler@va.wagner.com The Daniel H. Wagner Prize for Excellence in Operations Research Practice emphasizes the quality and coherence of the analysis used in practice. Dr. Wagner strove for strong mathematics applied to practical problems, supported by clear and intelligible writing. This prize recognizes those principles by emphasizing good writing, strong analytical content, and verifiable practice successes Keynote Tuesday Davidson Ballroom C-MCC Edelman Reprise-UPS Keynote Session Chair: Michael Trick, Carnegie Mellon University, Pittsburgh, PA, (IFORS President) trick@cmu.edu 1 - UPS Optimizes Delivery Routes Jack Levis, Director of Process Management, UPS, Timonium, MD, United States, jlevis@ups.com, Ranganeth Nuggehalli UPS, the leading logistics provider in the world, and long known for its penchant for efficiency, embarked on a journey to streamline and modernize its pickup and delivery operations in 2003. This journey resulted in a suite of systems, including an optimization system, which is called “On Road Integrated Optimization and Navigation” (ORION). Every day, ORION provides an optimized route for each of UPS’ 55,000 U.S. drivers based on the packages to be picked up and delivered on that day. The innovative system creates routes that maintain the desired level of consistency from day to day. To bring this transformational system from concept to reality, UPS instituted extensive change management practices to ensure buy- in from both users and executives. Costing more than $250 million to build and deploy, ORION is expected to save UPS $300 to $400 million annually. ORION is also contributing the sustainability efforts of UPS by reducing the CO2 emissions by 100,000 tons annually. By providing a foundation for a new generation of advanced planning systems, ORION is transforming the pickup and delivery operations at UPS. Keynote Tuesday

Chair: Mariya Naumova, Rutgers University, 640 Bartholomes Road, Piscataway, NJ, 08854, United States, mnaumova@rci.rutgers.edu 1 - A Fast Algorithm For Discretization In A Big Data Space Abdelaziz Berrado, Associate Professor, University Mohammed V in Rabat, BP 765, Avenue Ibn Sina, Agdal, Rabat, 10080, Morocco, berrado@emi.ac.ma Discretizing continuous attributes is a necessary preprocessing step before association rules mining or using several inductive learning algorithms with a heterogeneous big data space. This important task should be carried out with a minimum information loss. We combine bagging and nonlinear optimization techniques to build an automated supervised, global and dynamic discretization algorithm that derives its ability in conserving the data properties from the Random Forest algorithm. Empirical results indicate the performance of our discretization algorithm. 2 - Orthogonal Tensor Decomposition Cun Mu, PhD Student, Columbia University, 500 West 120th Street, Room 315, New York, NY, 10027, United States, cm3052@columbia.edu, Donald Goldfarb Many idealized problems in signal processing, machine learning, and statistics can be reduced to the problem of finding the symmetric canonical decomposition of an underlying symmetric and orthogonally decomposable (SOD) tensor. In this talk, we will address several practical issues arising from conducting this orthogonal tensor decomposition. 3 - Distance-based Methods For Classification Of Groups Of Objects Mariya Naumova, Rutgers University, 110 Frelinghuysen Rd., Piscataway, NJ, 08854, United States, mnaumova@rci.rutgers.edu Given a finite number of learning samples from several populations (groups) and a collection of samples from the union of these populations, it is required to classify the entire collection (not a single sample) to one of the groups. Such problems often arise in medical, chemical, biological and technical diagnostics, classification of signals, etc. We consider different methods of solving the problem based on distance formulas and make comparison of their quality based on numerical results. We give an illustrative example with real data to demonstrate the effectiveness of the classification methods. TD02 101B-MCC Data Mining in Medical and Brain Informatics II Sponsored: Data Mining Sponsored Session Chair: Chun-An Chou, SUNY Binghamton, 4400 Vestal Parkway East, Binghamton, NY, 13902, United States, cachou@binghamton.edu Co-Chair: Sina Khanmohammadi, SUNY Binghamton, 4400 Vestal Parkway East, Binghamton, NY, 13902, United States, skhanmo1@binghamton.edu 1 - Statistical Learning Of Neuronal Functional Connectivity Chunming Zhang, University of Wisconsin - Madison, cmzhang@stat.wisc.edu Identifying the network structure of a neuron ensemble is critical for understanding how information is transferred within such a neural population. We propose a SIE regularization method for estimating the conditional intensities under the GLM framework to better capture the functional connectivity among neurons. A new algorithm is developed to efficiently handle the complex penalty in the SIE-GLM for large sparse data sets applicable to spike train data. Simulation results indicate that our proposed method outperforms existing approaches. An application of the proposed method to a real spike train data set provides some insight into the neuronal network.

333