Our theme this year: Moving from Sick-care to Healthcare -Interdisciplinary collaboration between medical and STEM fields.

Agenda Will Include:

• Rapid-fire Presentations
• Poster Session (Submit Your Abstract to Participate by Nov 4th)
• Panel Session 

Featuring

• Tarun Mohan Lal, Chief Analytics Officer and Enterprise Vice President | Navicent Health
• Joyce Siegele, Director of Productivity Improvement | Northside Hospital 
•  Greg Esper, Vice Chair, Clinical Affairs, Department of Neurology | Emory University School of Medicine
• Victoria Jordan, Vice President, Quality | Emory Healthcare 

Representatives from the humanitarian sector, government, NGOs, foundations and private industry, and academia present diverse perspectives in health and humanitarian challenges through keynote addresses, panel discussions, focused workshops, lunchtime group discussions, and interactive poster sessions covering a broad set of research topics and applications.

The conference is sure to bring together an abundance of professionals active in the global health and humanitarian sectors from around the world. The event is chaired annually by the Georgia Tech Center for Health & Humanitarian Systems (CHHS), NCSU, INSEAD Humanitarian Research Group, MIT Humanitarian Response Lab, Northeastern University. This year’s conference is pleased to have as Co- organizers The International Association of Public Health Logisticians (IAPHL) and People that Deliver, and University of Rwanda Regional Center for Excellence as our Host.

We invite you to attend and particpate as a presenter in the following areas: collaborative workshops, oral presentations and poster sessions. To see our requirements and submit a proposal visit our Call For Presentations Page

1. Improving Consumer Welfare and Manufacturer Profit via Government Subsidy Programs: Subsidizing Consumers or Manufacturers?

Most consumers in rural areas of many developing countries cannot afford to purchase certain livelihood improvement products such as home appliances. To improve consumer welfare and manufacturer profit, many governments launch different types of subsidy programs that offer subsidies to consumers, manufacturers, or both. Motivated by a subsidy program developed by the Chinese government in 2007, we present a parsimonious model to determine the optimal subsidy program in different settings so as to gain a better understanding about the conditions under which it is optimal for the government to subsidize consumers only, manufacturers only, or both. Our analysis reveals that the structure of the optimal subsidy program depends on (a) whether there is a well- established market selling price for the products; and (b) the relative emphasis that the government places on consumer welfare versus manufacturer profit. Also, we find that governments can improve consumer welfare by developing subsidy programs that involve competing manufacturers with different market sizes and adequate capacities. Our findings provide insights for developing effective government subsidy programs.

2. A Balancing Act of Regulating On-demand Ride Services

Regulating on-demand ride-hailing services (e.g., Uber and DiDi) requires a balance of multiple competing objectives: encouraging innovative business models (e.g., DiDi), sustaining traditional industries (e.g., taxi), creating new jobs, and reducing traffic congestion. This study is motivated by a regulatory policy implemented by the Chinese government in 2017 and a similar policy approved by the New York City Council in 2018 that regulate the “maximum” number of registered Uber/DiDi drivers. We examine the impact of these policies on the welfare of different stakeholders (i.e., consumers, taxi drivers, on-demand ride service company, and independent drivers). By analyzing a two-period dynamic game that involves these stakeholders, we find that without government intervention, the on-demand ride service platform can drive the traditional taxi industry out of the market under certain conditions. Relative to no regulations and a complete ban policy, a carefully designed regulatory policy can strike a better balance of multiple competing objectives. Finally, if a government can reform the taxi industry by adjusting the taxi fare, then lowering the taxi fare instead of imposing a strict policy toward on-demand ride services can improve the total social welfare.

Bio:

Dr. Jiayi Joey Yu is now a postdoctoral researcher at Department of IEOR, University of California, Berkeley. Her research interests have been focusing on practical problems arising from public policy issues, socially responsible operations and innovative business models. Dr. Yu received both Ph.D. and Bachelor degree from Department of Industrial Engineering, Tsinghua University and was a visiting Ph.D. student of UCLA Anderson School of Management. So far, she has published papers in journals like Management Science, Manufacturing & Service Operations Management, Decision Sciences, etc.  

Abstract:

In radiation therapy treatment planning optimization, selecting a set of clinical objectives that are tractable and parsimonious yet clinically effective is a challenging task. In clinical practice, this is typically done by trial and error based on the treatment planner's subjective assessment, which often makes the planning process inefficient and inconsistent. We develop the objective selection problem that infers a sparse set of objectives for prostate cancer treatment planning based on historical treatment data. We formulate the problem as a non-convex bilevel mixed-integer program using inverse optimization and highlight its connection with feature selection to propose greedy heuristics as well as application-specific methods that utilize anatomical information of the  patients. Our results show that the proposed heuristics find objectives that are near optimal. Using curve analysis for dose-volume histograms, we show that the learned objectives closely represent latent clinical preferences by recovering historical treatment for each patient.

Bio:

Tayo is a fifth-year PhD candidate at Rice University in the Department of Computational and Applied Mathematics. Tayo's research interests include integer programming theory and healthcare applications, particularly in cancer treatment. He is a Visiting Graduate Student at The University of Texas MD Anderson Cancer Center in the Department of Radiation Oncology.

Concussion, the most common type of traumatic brain injury, has been identified as a public health issue. Recent research has exposed a troubling relationship between concussion and long-term health consequences to brain health, such as cognitive impairment, depression, and neurodegenerative disease. Appropriate concussion management is thought to play a critical role in improving long and short-term health outcomes for those with concussion. Unfortunately, the diagnosis and post-injury management of concussion remains challenging for many reasons, including: the lack of a gold standard diagnostic marker, the potential for strategic symptom reporting, and the need for guidelines built on rigorous analysis of large, observational clinical datasets. In this research, we develop two frameworks to address these issues. In the first framework, we formulate a multi-agent Partially Observable Markov Decision Process (mPOMDP) to model both the patient’s and doctor’s perspectives in sequential treatment decision problems. We analyze the role of strategic symptom-reporting on the optimal timing of return-to-play from sports-related concussion. While classical results for POMDPs do not hold for the mPOMDP, we derive conditions which ensure that the doctor’s optimal policy follows a threshold-type structure. In the second framework, we formulate the two-threshold problem (TTP) as a stochastic programming model to determine which patients should be diagnosed as positive, negative, or deferred due to a lack of conclusive evidence. We characterize the optimal solution to TTP and develop data-driven methodologies to solve and calibrate TTP. For both frameworks, we conduct numerical studies using multi-center data from the CARE Consortium – the largest available dataset on sports-related concussion. We show that our frameworks outperform existing methods commonly used in practice and use our findings to generate clinical insights for concussion management. The models developed in this research provide technical contributions to data-driven decision-making that can be applied broadly to other areas within and beyond healthcare.

Bio:

Gian-Gabriel Garcia is a PhD Candidate in the Industrial and Operations Engineering Department at the University of Michigan. He holds a Bachelor’s degree in Industrial Engineering from the University of Pittsburgh and a Master’s degree in Industrial and Operations Engineering from the University of Michigan. In his research, Gian is interested in developing data-driven frameworks for predictive and prescriptive analytics as motivated by high-impact problems in healthcare.  His current research focuses on (1) using large clinical datasets to gain patient-specific insights on disease progression and (2) combining these insights with stakeholders’ perspectives to improve diagnosis and treatment decisions. His research has been applied to concussion, glaucoma, and cardiovascular disease. It has been recognized by the National Science Foundation Graduate Research Fellowship, the INFORMS Bonder Scholarship for Applied Operations Research in Health Services, the Rackham Merit Fellowship, the SMDM Lee B. Lusted Prize in Quantitative Methods and Theoretical Developments, and first prize at the INFORMS Minority Issues Forum Poster Competition.

We developed a framework which enables one to strengthen low-level RLT relaxations by adding cutting planes, rather than additional variables. Our framework allows to generate cutting planes at any given level of the hierarchy, by optimally weakening linear inequalities that are valid at any given higher level. The framework is applicable to mixed 0-1 polynomial programs with bounded continuous variables. It consists of three main components. First, we give procedures for constructing low-degree polynomial over- and under-estimators of polynomials of higher degree. Second, we show how to use those over- and under-estimators to convert any valid linear inequality for a given high-level RLT relaxation into an exponentially large collection of cutting planes for a given low-level RLT relaxation. Third, we present separation algorithms that, under certain conditions, enable one to find in polynomial time the most-violated cutting plane in a given collection.

In order to explore the potential of these new cutting planes, we will present some computational results, using the quadratic knapsack problem (QKP) as an example. It turns out that, for the difficult sparse QKP instances, our cutting planes close around half of the gap between the upper bound from the standard first-level relaxation and the optimum.

Bio:
Dr Franklin Djeumou Fomeni is a Postdoctoral Research Fellow at the Inter-University Research Centre on Enterprise Networks, Logistics and Transportation (CIRRELT) in Montreal, Canada. He received a B.Sc. and a Masters in Mathematics from the University of Yaounde 1, Cameroon. He was then selected by the African Institute for Mathematical Sciences (AIMS, www.nexteinstein.org) as one of the 50 brightest Africa-wide students to complete their 10-month Postgraduate Diploma program in Mathematical Sciences in Cape Town, South Africa. He subsequently completed an MSc in Applied Mathematics at the University of the Witwatersrand, South Africa. Franklin obtained his PhD in Operations Research from Lancaster University, United Kingdom, under the supervision of Adam Letchford.  He has received several awards for his academic achievements, among which, the Kingsman Prize for the best PhD thesis in Operations Research at Lancaster University, UK; the Rand Merchant Gold Medal, awarded to the best MSc student within the Faculty of Science at the University of the Witwatersrand, South Africa.

Franklin held postdoctoral positions at Polytechnique Montreal, Canada and at the Centre for Transport and Logistics (CENTRAL) at Lancaster University, UK.  His primary research interest is in Combinatorial Optimization, where he develops exact and heuristic solution methods using techniques such as Cutting planes, Branch-and-Cut, Dynamic Programming. He also has a particular interest in the application of optimization to problems in energy, transportation networks, Logistics, air traffic flow management and Manufacturing. His research has been published in journals such as Mathematical Programming, Transportations Research Part B: Methodological, European Journal of Operational Research, Computers & Operations Research and INFORMS Journal on Computing.

She is the Distinguished Professor of Engineering in the Electrical Engineering and Computer Science (EECS) Department at the Massachusetts Institute of Technology. She is also the department head of EECS. Her research expertise includes optimization theory, with emphasis on nonlinear programming and convex analysis, game theory, with applications in communication, social, and economic networks, distributed optimization and control, and network analysis with special emphasis on contagious processes, systemic risk and dynamic control.

Professor Ozdaglar is the recipient of a Microsoft fellowship, the MIT Graduate Student Council Teaching award, the NSF Career award, the 2008 Donald P. Eckman award of the American Automatic Control Council, the Class of 1943 Career Development Chair, the inaugural Steven and Renee Innovation Fellowship, and the 2014 Spira teaching award. She served on the Board of Governors of the Control System Society in 2010 and was an associate editor for IEEE Transactions on Automatic Control. She was the inaugural area co-editor for the area entitled "Games, Information and Networks” in the journal Operations Research. She is the co-author of the book entitled “Convex Analysis and Optimization” (Athena Scientific, 2003).

Some Challenges in the Design and Application of Simulator
Experiments to Biomechanics and Biomaterials Research

Abstract:

This talk will demonstrate how simulator experiments can be combined
with physical experiments to solve problems of Biomechanics and the tissue
engineering of Biomaterials. Biomechanics applies mechanical engineering
principles to understand the movements of living things. The use of statistical
methods to help devise treatments of anomalies in the movements of living
systems will be described. Three classes of problems will be introduced to
illustrate this interdisciplinary Biomechanics/Statistical approach. We will
consider (1) the design of prosthetic joints, (2) the identification of the root
causes of prosthetic joint failures, and (3) the development of replacement
meniscal tissue.

Bio:

Professor Santner has contributed to the design and analysis of Selection and Ranking experiments, to the analysis of Discrete Data and most recently to the design and analysis of experiments that use deterministic computer simulator as research platforms. He has been a visiting professor at Purdue, Duke University, the University of Washington, and the University of Munich. Santner is a Fellow of the American Statistical Association, the Institute of Mathematical Statistics, and the American Association for the Advancement of Science.  Today's talk will describe several interdisciplinary problems which are at the confluence of Biomechanical Engineering and the Statistical analysis of finite element model data.   

Smaller p-values and shorter confidence intervals via information sharing

Abstract:

Mixed effects models are used routinely in the biological and social sciences to share information across groups and to account for data dependence. The statistical properties of procedures derived from these models are often quite good on average across groups, but may be poor for any specific group. For example, commonly-used confidence interval procedures may maintain a target coverage rate on average across groups, but have near zero coverage rate for a group that differs substantially from the others. In this talk we discuss new confidence interval and p-value procedures that maintain group-specific frequentist guarantees, while still sharing information across groups to improve precision and power.

Bio:

Peter Hoff is Professor of Statistical Science at Duke University. His current research interests include multivariate statistics and hierarchical modeling. He is a Fellow of the ASA and IMS, and is the author of “A First Course in Bayesian Statistical Methods”.

Modelling Moving Points: Identifying Ship Trajectories and Optimizing Photon Strategy

Abstract: 

Research of movement trajectory is undergoing a revolution due to the growing volumes of high-resolution spatio-temporal movement data and models . A key challenge now is the analysis of massive movement datasets with largely different sampling rates and accuracy such as automatic identification system (AIS) geolocation data. In particular, it is important and challenging to identify movement trajectories which have close paths and changing speeds and angles particularly in an unsupervised manner. In this talk, I will talk about our methods for modelling the challenging AIS data and optimizing photon strategy for medical images.

Bio:

Dr. Hsin-Hsiung "Bill" Huang is an Assistant Professor and the Associate Chair in the Department of Statistics and Data Science at University of Central Florida (UCF). Dr. Huang received his Ph.D. in Statistics from the University of Illinois at Chicago and two MS degrees from the Georgia Institute of Technology and National Taiwan University as well as BA and BS from National Taiwan University. His scholarly interests and expertise include clustering, classification, dimension reduction. His research addresses challenges in analyzing big data, interdisciplinary research, and developing new statistical methods. His current research is partially sponsored by an NSF grant and an NIH grant.

Neural network estimation of likelihood ratios for testing and detection

Abstract: 

Various problems in Statistics and Engineering require the computation of the likelihood ratio function of two probability densities. In classical approaches the two densities are assumed known or to belong to some known parametric family. We replace this requirement with the need for availability of data sampled from the densities of interest. For most well known problems in Detection and Hypothesis testing we develop data-driven solutions by providing neural network based estimates of the likelihood ratio or its transformations. This task requires the definition of proper optimizations which can be used for the training of the neural networks. The main purpose of this work is to offer a simple and unified methodology for defining such optimization problems with guarantees that the solution is indeed the desired function. Our results are extended to cover estimates of likelihood ratios of conditional densities and estimates of statistics encountered in local approaches. 

Short Bio:

George V. Moustakides received the diploma in Electrical and Mechanical Engineering from the National Technical University of Athens, Greece in 1979, the MSE in Systems Engineering from the University of Pennsylvania in 1980, and the M.Sc and Ph.D in Electrical Engineering and Computer Science from Princeton University in 1983. Since 1988 he is with the University of Patras, Greece initially with the Computer Engineering and Informatics department and later with the department of Electrical and Computer Engineering. In 2017 he also joined the Computer Science department at Rutgers University. In the past he held various visiting or long-term appointments with INRIA France, Princeton University, University of Pennsylvania, Columbia University, University of Maryland, Georgia Institute of Technology, University of Southern California, University of Illinois at Urbana-Champaign and Rutgers University. His interests include Sequential Detection, Statistical Signal Processing and Machine Learning. From 2011 to 2014 and from 2016 to 2018 he served as Associate Editor for the IEEE Transactions on Information Theory.

BET on Independence

Abstract: 

We study the problem of nonparametric dependence detection. Many existing methods may suffer severe power loss due to non-uniform consistency, which we illustrate with a paradox. To avoid such power loss, we approach the nonparametric test of independence through the new framework of binary expansion statistics (BEStat) and binary expansion testing (BET), which examine dependence through a novel binary expansion filtration approximation of the copula. Through a Hadamard transform, we find that the symmetry statistics in the filtration are complete sufficient statistics for dependence. These statistics are also uncorrelated under the null. By utilizing symmetry statistics, the BET avoids the problem of non-uniform consistency and improves upon a wide class of commonly used methods (a) by achieving the minimax rate in sample size requirement for reliable power and (b) by providing clear interpretations of global relationships upon rejection of independence. The binary expansion approach also connects the symmetry statistics with the current computing system to facilitate efficient bitwise implementation. We illustrate the BET with a study of the distribution of stars in the night sky and with an exploratory data analysis of the TCGA breast cancer data.

Bio:

Kai Zhang is currently an associate professor with tenure at the Department of Statistics and Operations Research, UNC Chapel Hill. Dr. Zhang obtained his bachelor’s degree from Peking University in 2003, his Ph.D. degree in mathematics from Temple University in 2007, and his Ph.D. degree in statistics from the Wharton School, University of Pennsylvania in 2012. His research interests include nonparametric statistics, high-dimensional statistics, and post-selection inference.

The workshop hopes to attract graduate and senior undergrad students in ECE, IE, and related disciplines both from Georgia Tech and nationwide. It consists of research talks, presentations, and panel discussions based on the following themes:

• Dealing with uncertainty in power systems: Renewable integration, market design;
• New ways to solve complex optimization problems in power system operations and planning;
• Data-analytics for power systems;
• How to promote collaborations between academia and energy industry.

Energy Systems and Optimization Workshop Website

Abstract:  QPLEX is a nonstationary, nonparametric, non-Markovian modeling and analysis paradigm for stochastic networks. QPLEX generates transient distributions of key performance metrics, such as the number of customers and the virtual waiting time at each station at all times. From a modeling perspective, QPLEX is quite versatile: for example, it can accommodate time-varying arrival processes, arbitrary time-varying service-time distributions, time-varying server counts, abandonments, balking, probabilistic routing and capacity/routing policies based on cycle-time distributions.

In this talk, we will present a proof-of-concept that is remarkably accurate, widely applicable, and extremely fast. We will describe the mechanisms underlying the QPLEX calculus as well as the principles of QPLEX modeling. We will also discuss future directions and prospective application areas.  This is joint work with Steve Hackman (Georgia Tech).

Bio: Ton Dieker is Associate Professor of Industrial Engineering and Operations Research at Columbia University, and a member of Columbia’s Data Science Institute. He received an M.Sc. from Vrije Universiteit Amsterdam (2002) and a Ph.D. from University of Amsterdam (2006). Prior to joining Columbia, he was the Fouts Family Associate Professor at Georgia Tech. Honors and awards include the Goldstine Fellowship from IBM Research, the Erlang Prize from the INFORMS Applied Probability Society, and a PECASE Award from the White House. He serves/has served on the editorial board of several journals in Operations Research and Applied Probability.

The French Consulate General in Atlanta, Emory University’s Rollins School of Public Health, and the Georgia Institute of Technology Center for Health and Humanitarian Systems will host a one-day forum uniting French and American NGOs and health experts. Over the course of two panel discussions, experts will touch on medical commodities, the use of pre-packed kits and the opportunity for collaboration between corporations and humanitarian organizations. They will also explore the connection between Atlanta-based programming and organizations with global humanitarian logistics programing.

SESSION OVERVIEW

Though building a Connected Intelligent Supply Chain can seem daunting, we say Think Big, Start Small, Move Fast. To that end, the Microsoft Azure ecosystem provides platforms, products, and services to accelerate incremental building of hybrid solutions no matter the scale. From Industrial IoT to Blockchain to Modern Workplace to Business Intelligence and Visualization to Augmented Reality, Microsoft spans the entire breadth and depth of digital transformation. Finally, layering AI into these solutions to further drive insight, safety, and value has never been easier than it is today on Azure with Cognitive Services and Azure Machine Learning.

SESSION SPEAKER

Dr. Keith Duggar, Principal Industry Technology Strategist at Microsoft Corporation, is a Georgia Tech and MIT alum who works with enterprise manufacturing companies to envision, architect, and execute digital transformation including all technological aspects of Connected Intelligent Enterprise. He brings 20+ Years of Quantitative Data Analytics, Machine Learning, and Applied Math experience and 10+ Years of High Performance Low Latency Decentralized Computing, Data Modeling, Data System Design, and Data Strategy experience.

Register Online for upcoming SCL IRC seminars

Attendance to the sessions is complimentary for SCL corporate partners, SCL Industry Advisory Board members, SCL affiliated faculty and students, and students enrolled in the Masters in Supply Chain Engineering program.

To take advantage of the included lunch, you must register by deadline noted on the registration page.

If you have any questions, please email event@scl.gatech.edu.

We strongly encourage students to act now to seek full-time employment, internships, and projects (rather than waiting until the end of the semester). Plus, enjoy a free pizza lunch!

EVENT DETAILS

Where: ISyE Main Bldg, 2nd Floor Atrium

When: Tuesday, November 5, 2019 | 11am - 2pm

What: The session will include:

• Networking opportunities in the ISyE atrium
• Food and refreshments

Please plan on staying for the duration of the event and bring copies of your resume and business cards. Dress is business casual.

REGISTER ONLINE by October 27th to upload your resume and have it forwarded to the organization representatives!

EVENT SPONSOR

The event is sponsored through the generosity and support of JP Morgan Chase & Co. and APICS - Atlanta Chapter. APICS is a non-profit educational organization addressing operations management and supply chain management issues, and providing professional development opportunities to its members. APICS Membership is free for full time students. Join today at www.apics.org/join and start networking at local APIC Atlanta events. Also make sure to stop by the APICS table at the event.

Abstract:

Medicine has evolved from one-treatment-fits-all to personalized medicine that takes inter-person difference into account for treatment. Recent development in cancer medicine is pushing this evolution to a new paradigm: heterogeneity exists not only between patients but also within each tumor. Intra-tumor molecular heterogeneity has been found to be a leading cause of treatment failure for some aggressive cancers. This urges the need for mapping out regional molecular status across each tumor. It is a challenging task because no single data source can accomplish this goal. Localized biopsy samples are scarce due to their invasive nature. Imaging portrays the entire tumor but does not provide direct measure of molecular characteristics. Also available are knowledge-based mechanistic models, which may be inaccurate or incomplete. Fusion of all these data/knowledge sources provides the best promise for mapping out intra-tumor molecular heterogeneity and ultimately enabling new treatment with unprecedented precision.  

In this talk, I will introduce a modeling framework allowing for knowledge-infused global (imaging)-local (biopsy) data fusion. This framework has roots in semi-supervised learning but innovates through mechanistic model integration and simultaneous feature & instance selection. An extension of this model will also be discussed, which allows for uncertainty quantification. Case studies of glioblastoma – the most aggressive type of brain cancer – are presented with real data, demonstrating the potentially transformative impact on treatment (surgical resection, radiation therapy, genetic therapy). 

Interestingly, we found that the need for knowledge-infused global-local data fusion exists beyond medicine, such as early detection of forest fires and mapping out regional poverty in the developing world. It will be interesting to test the utility of the proposed methodology in these applications.

Bio:

Dr. Jing Li is an Associate Professor in Industrial Engineering & Computer Engineering at Arizona State University (https://www.public.asu.edu/~jli09/). She received her B.S. from Tsinghua University, and an M.A. in Statistics and a Ph.D. in Industrial and Operations Engineering from the University of Michigan. Her research interests are data fusion and statistical machine learning intersecting with health care domains having complex data structures.  Dr. Li’s research is sponsored by NIH, NSF, DOD, Arizona State, Mayo Clinic, and biomedical industry. She co-founded the ASU-Mayo Clinic Center for Innovative Imaging, conducting various collaborative projects with the Departments of Radiology, Neurology, Neurosurgery, and Radiation Oncology at Mayo Clinic. She is an NSF CAREER awardee, a recipient of a Best Paper Award and a Best Application Paper Award from IISE Transactions, a recipient of the Harold Wolff-John Graham Award (Best Paper) from the American Academy of Neurology, and a recipient of the Harold G. Wolff Lecture Award (Best Paper) by the American Headache Society. She is a former Chair for the Data Mining Subdivision of INFORMS. She is currently the Editor-in-Chief for Quality Technology and Quantitative Management, an Associate Editor for IISE Transactions on Healthcare Systems Engineering, and on the editorial board of Journal of Quality Technology. 

This accelerated learning boot camp covers 2.5 days of instructor-led facilitation and is designed to cover key learning points of the CSCP learning system. The learning system content is included in the course. 

This boot camp is an exam preparatory course for the self-study learner planning to take the APICS CSCP certification exam. The course accelerates learning by focusing on key concepts and terminology covered in the exam. This boot camp, in tandem with the reading materials and online learning courseware, provides a roadmap for those who intend to sit for the exam.

The APICS Certified Supply Chain Professional (CSCP) program takes a broad view of the field, extending beyond internal operations to encompass all the steps throughout the supply chain--from the supplier, through the company, to the end consumer--and how to effectively manage the integration of these activities to maximize the company’s value chain.

The CSCP program provides candidates with the necessary tools to effectively manage global supply chain activities and enables them to implement best practice approaches to increase supply chain efficiencies.

WHO SHOULD ATTEND

The APICS CSCP is designed for supply chain professionals and people managing and planning extended supply chains especially people in:

• Supply chain design
• Supplier management
• Transportation
• Supply chain management
• Distribution channels
• 3PL, 4PL management
• Customer management
• Supply chain consulting

HOW YOU WILL BENEFIT

• Master advanced supply chain management principles that extend beyond an organization’s internal end-to-end operations from suppliers to customers
• Apply supply chain knowledge and analytical skills to streamline operations and produce bottom line results
• Explain the role of each element of the integrated supply chain concept
• Understand how successful supply chain management adds value to your organization
• Learn how to develop a supply chain strategy that aligns with corporate strategy
• Understand natural dynamics within the supply chain to optimize performance and profitability
• Evaluate the process constraints and choices within Global Logistics to establish a plan linked to overall strategy
• Effectively use customer data to improve service performance and increase value to suppliers and customers
• Understand the strategic importance of purchasing and supply relationships
• Understand the innovative technologies enabling collaborative commerce and global visibility
• Apply technology to enhance performance of distribution, reverse logistics, and global supply chain communications

WHAT IS COVERED

Supply Chain Design

• Addresses the concepts and strategies used for developing a supply chain strategy aligning with business goals and corporate strategy
• Designing a supply chain for the flow of product, information and cash
• Understanding and using the SCORreg model

Supply Chain Planning and Execution

• Addresses the processes required to procure and deliver goods and services
• Management of demand and supply relationships

Supply Chain Planning and Execution

• Related concepts of order management and customer service
• Defining and measuring logistics customer service outputs

Supply Chain Improvement and Best Practices

• Managing and balancing supply and demand through measuring, analyzing and improving supply chain processes
• Compliance with standards and regulations
• The importance of sustainable best practices and social responsibility
• Assessment of risk within the supply chain

This course prepares students in the basics of transportation operations and analysis.  The course includes review of the key elements of transportation such as: modes of transportation, transportation procurement, cost minimization techniques, the role of ports in global logistics, and international trade terms.  The course also will discuss emerging trends in North American transportation markets, emerging techniques, and greenhouse gas emissions reduction.

WHO SHOULD ATTEND

This course is designed for Supply Chain Managers, Distribution Managers, Transportation Planners, Transportation Clerks, Transportation Analysts, and Transportation Managers and learners seeking to enter these roles.  Supply chain professionals from other domains will also benefit through gaining insights into transportation operations.

HOW YOU WILL BENEFIT

Upon completion of this course, you will be able to:

• Understand the characteristics and best use of specific transportation modes
• Apply transportation cost analysis techniques
• Understand the multimodal role of ports in global trade
• Identify and apply best practices in transportation procurement
• Understand how to minimize transportation costs through consolidation techniques
• Understand the role of Incoterms in global trade
• Understand emerging techniques in logistics including techniques to reduce greenhouse gas emissions in logistics

WHAT IS COVERED

• Comparison of characteristics of alternative transportation modes
• Components of Port Logistics systems
• Best practices in transportation procurement
• Application of practical transportation cost analysis techniques
• INCOTERMS purpose, types, and use
• Greenhouse gas emission generation in logistics and mitigation strategies
• New business models in logistics enabled by emerging technologies

Numerous tactical decisions must be made in the midst of public health and humanitarian challenges. Many of these decisions are concerned with the timely and efficient procurement, allocation, and distribution of resources (e.g. funds, supplies, volunteers) through a supply chain. This course will explore methodologies for tactical decision making including procurement and inventory policies, strategies for distribution and allocation of limited resources, and transportation decisions.

WHO SHOULD ATTEND

This course is designed for tactical and strategic members of non-governmental organizations, private corporations involved in the humanitarian relief efforts, U.S. federal government employees, U.S. state or local government employees, humanitarian foundation members, program managers, directors of field operations, disaster relief managers, director of logistics and inventory, and logistics officers.

HOW YOU WILL BENEFIT

• Provide immediate impact to your organization through applied and real-world case studies.
• Learn to bring NGOs, businesses, or government entities together to enhance collaboration, cooperation, and communication.
• Discover current trends and procedures to help your organization and team members get and stay ahead of the curve.
• Build a critical knowledge base to make tactical decisions around inventory, routing, and distribution.
• Deliver best practices to measure and evaluate outcomes of humanitarian relief efforts.
• Transform the humanitarian sector with increased capacity to participate in planning and strategic decision-making for effective supply-chain management.

WHAT IS COVERED

• Procurement decisions
• Inventory techniques for a single event versus ongoing operations
• Resource allocation trade-offs
• Resource allocation and transportation strategies
• Fund allocation strategies

ON-CAMPUS COURSE MATERIALS

Provided

• Readings, case studies, spreadsheet files and lecture slides

Recommended

• Laptop with Microsoft Excel and wi-fi capability

COURSE PREREQUISITES

No prerequisites.

CERTIFICATE INFORMATION

This course is part of the Supply Chain Management (SCM) Certificate and Health and Humanitarian Supply Chain Management Certificate.

PROGRAM TIMES

On the first day, please check in at least 30 minutes before the class start time.

• First Day (9:00am-5:00pm)
• Second Day (9:00am-5:00pm)

Often the lack of cooperation and coordination between organizations or stakeholders lead to inefficiencies, despite having common goals. A systems view is needed to ensure appropriate use of scarce resources to meet the multiple, and often conflicting, short- and long-term goals from multiple constituents. This course will focus on conceptual and modeling skills to understand and effectively manage supply chains and operations from a systems perspective. Models will address system characteristics (e.g., demand dependencies) that drive system dynamics and policies to regulate performance. Course topics include methods for improving coordination and collaboration, addressing demand dependencies, and reliably measuring and evaluating system performance.

WHO SHOULD ATTEND

This course is designed for tactical and strategic members of non-governmental organizations, private corporations involved in the humanitarian relief efforts, U.S. federal government employees, U.S. state or local government employees, humanitarian foundation members, program managers, directors of field operations, disaster relief managers, director of logistics and inventory, and logistics officers.

HOW YOU WILL BENEFIT

• Identify opportunities for coordination within organizations and collaboration across organizations for increased efficiency and improved outcomes.
• Describe the strategic behavior of decision-makers and the impact of the market (or contract) structure on the participant's actions and the overall system dynamics.
• Define evaluation metrics in alignment with the system goals and structure system operations and incentives that address and evaluate these metrics.

WHAT IS COVERED

• Coordination and collaboration improvement
• Demand dependencies
• Inventory variability and costs mitigation and reduction

COURSE MATERIALS

Provided

• Participants receive readings, case studies, spreadsheet files, and lecture slides, and will be given access to the pre-course web-based activities.

Recommended

• Students need a laptop with Microsoft Excel and the ability to connect to a high-speed internet connection (internet access is provided for onsite portions of course).

COURSE PREREQUISITES

For those interested in earning the Health and Humanitarian Supply Chain Management Certificate, this course is the third and final of the three-course certificate program. To earn the certificate, participants must register and complete the following courses within three years:

1. Pre-planning Strategy for Health and Humanitarian Organizations
2. Tactical Decision Making in Public Health and Humanitarian Response
3. Systems Operations and Strategic Interactions in Health and Humanitarian Supply Chains

Delivering sustainable solutions for public health and humanitarian events is generally impacted by dynamic and hard-to-measure factors such as geographic location, social and political conditions, as well as local economy and infrastructure. Mapping long-term development efforts and responding to healthcare or humanitarian challenges often involves uncertainty in timing, scope, or scale. Pre-planning for these efforts requires an understanding of various topics such as forecasting, distribution, network design, and inventory management. This course will examine methods and models for making pre-planning decisions and explore the significant value that is obtained through informed decision-making in advance of an unpredictable event or long-term strategy for sustaining wellness.

WHO SHOULD ATTEND

This course is designed for tactical and strategic members of non-governmental organizations and private corporations involved in the humanitarian relief efforts, as well as U.S. federal, state and local government employees; humanitarian foundation members; program managers; directors of field operations; disaster relief managers; directors of logistics and inventory; and logistics officers.

HOW YOU WILL BENEFIT

• Provide immediate impact to your organization through learnings gained from applied and real-world case studies.
• Learn to bring NGOs, businesses, or government entities together to enhance collaboration, cooperation, and communication.
• Discover current trends and procedures to help organizations and team members get and stay ahead of the curve.
• Build a critical knowledge base to make tactical decisions around inventory, routing, and distribution.
• Deliver best practices to measure and evaluate outcomes of humanitarian relief efforts.
• Transform the humanitarian sector with increased capacity to participate in planning and strategic decision-making for effective supply-chain management.

WHAT IS COVERED

• Analytical models
• A network approach
• Forecasting techniques
• Uncertain events strategies

COURSE MATERIALS

Provided

• Participants receive readings, case studies, spreadsheet files, and lecture slides, and will be given access to the pre-course web-based activities.

Recommended

• Students need a laptop with Microsoft Excel and the ability to connect to a high-speed internet connection (internet access is provided for onsite portions of course).

STUDENT REQUIREMENTS

Students need a laptop with Microsoft Excel and the ability to connect to a high-speed internet connection (internet access is provided for onsite portions of course).

COURSE PREREQUISITES

For those interested in earning the Health and Humanitarian Supply Chain Management Certificate, this course is the first of the three-course certificate program. To earn the certificate, participants must register and complete the following courses within three years:

1. Pre-planning Strategy for Health and Humanitarian Organizations
2. Tactical Decision Making in Public Health and Humanitarian Response
3. Systems Operations and Strategic Interactions in Health and Humanitarian Supply Chains

AutoAssist: A Framework to Accelerate Training of Deep Neural Networks

Abstract:

Deep neural networks have yielded superior performance in many contemporary applications. However, the gradient computation in a deep model with millions of instances leads to a lengthy training process even with modern GPU/TPU hardware acceleration. In this talk, I will propose AutoAssist, a simple framework to acceler- ate training of a deep neural network. Typically, as the training procedure evolves, the amount of improvement by a stochastic gradient update varies dynamically with the choice of instances in the mini-batch. In AutoAssist, we utilize this fact and design an instance shrinking operation that is used to filter out instances with relatively low marginal improvement to the current model; thus the computationally intensive gradient computations are performed on informative instances as much as possible. Specifically, we train a very lightweight Assistant model jointly with the original deep network, which we refer to as the Boss. The Assistant model is designed to gauge the importance of a given instance with respect to the current Boss such that the shrinking operation can be applied in the batch generator. With careful design, we train the Boss and Assistant in a nonblocking and asynchronous fashion such that overhead is minimal. To demonstrate the effectiveness of AutoAssist, we conduct experiments on two contemporary applications: image classification using ResNets with varied number of layers, and neural machine translation using LSTMs, ConvS2S and Transformer models. For each application, we verify that AutoAssist leads to significant reduction in training time; in particular, 30% to 40% of the total operation count can be reduced which leads to faster convergence and a corresponding decrease in training time.

Bio:

Inderjit Dhillon is the Gottesman Family Centennial Professor of Computer Science and Mathematics at UT Austin, where he is also the Director of the ICES Center for Big Data Analytics. Currently he is on leave from UT Austin and heads the Amazon Research Lab in Berkeley, California, where he is developing and deploying state-of-the-art machine learning methods for Amazon Search. His main research interests are in big data, deep learning, machine learning, network analysis, linear algebra and optimization. He received his B.Tech. degree from IIT Bombay, and Ph.D. from UC Berkeley. Inderjit has received several awards, including the ICES Distinguished Research Award, the SIAM Outstanding Paper Prize, the Moncrief Grand Challenge Award, the SIAM Linear Algebra Prize, the University Research Excellence Award, and the NSF Career Award. He has published over 200 journal and conference papers, and has served on the Editorial Board of the Journal of Machine Learning Research, the IEEE Transactions of Pattern Analysis and Machine Intelligence, Foundations and Trends in Machine Learning and the SIAM Journal for Matrix Analysis and Applications. Inderjit is an ACM Fellow, an IEEE Fellow, a SIAM Fellow and an AAAS Fellow.

Digital Transformation of Last-Mile Delivery

Don Ratliff, Regents' Professor Emeritus and Co-Executive Director, Georgia Tech Panama Logistics Innovation & Research Center

Last-mile delivery (i.e., the delivery of merchandise to a store or end user), is a prime target for digital transformation.  It is the most expensive and fastest growing leg of the supply chain and often the most critical to customer satisfaction.  Unfortunately, it is also the most complex leg of the supply chain.  We will discuss the delivery characteristics and trends that make digital transformation a necessity.  We will also discuss the five stages of digital transformation necessary to optimize delivery systems: 1. Data digitization; 2. Business intelligence; 3. Predictive analytics; 4. Automated delivery planning; and Adaptive management.  Finally, we will examine the advances in technology that enable transformation of last-mile delivery.

Register Online for upcoming SCL IRC seminars

Attendance to the sessions is complimentary for SCL corporate partners, SCL Industry Advisory Board members, SCL affiliated faculty and students, and students enrolled in the Masters in Supply Chain Engineering program.

To take advantage of the included lunch, you must register by deadline noted on the registration page.

If you have any questions, please email event@scl.gatech.edu.

Cost Learning via Inverse Optimal Transport

Abstract:

We propose a unified data-driven framework based on inverse optimal transport that can learn adaptive, nonlinear interaction cost function from noisy and incomplete empirical matching matrix and predict new matching in various matching contexts. We emphasize that the discrete optimal transport plays the role of a variational principle which gives rise to an optimization-based framework for modeling the observed empirical matching data. Our formulation leads to a bi-level non-convex optimization problem, for which we develop an efficient algorithm based on a saddle-point formulation of the optimization problem. The proposed approach has wide applicability including predicting matching in online dating, labor market, college application, and crowdsourcing. We back up our claims with numerical experiments on both synthetic data and real-world data sets.

Biography:

Dr. Xiaojing Ye is an associate professor at the Department of Mathematics and Statistics in Georgia State University, Atlanta, USA. Prior to joining Georgia State University in 2013, Dr. Ye was a visiting assistant professor at the School of Mathematics in Georgia Institute of Technology, USA. Dr. Ye received his doctoral degree in mathematics and master's degree in statistics from the University of Florida, USA in 2011 and 2009 respectively, and the bachelor’s degree in mathematics from Peking University in 2006. His research focuses on applied and computational mathematics, particularly PDE-based image analysis, numerical optimization, analysis and computational methods in machine learning.

More than 50 years of development have made mixed integer linear
programming (MILP) an extremely successful tool. MILP’s modeling
flexibility allows it describe a wide range of business, engineering
and scientific problems, and, while MILP is NP-hard, many of these
problems are routinely solved in practice thanks to state-of-the-art
solvers that nearly double their machine-independent speeds every
year. Inspired by this success, the last decade has seen a surge of
activity on the solution and application of mixed integer convex
programming (MICP), which extends MILP’s versatility by allowing the
use of convex constraints in addition to linear inequalities. In this
talk we cover various recent developments concerning theory,
algorithms and computation for MICP. Solvers for MICP can be
significantly more effective than those for more general non-convex
optimization, so one of the questions we cover in this talk is what
classes of non-convex constraints can be modeled through MICP. We also
cover various topics concerning the modeling and computational
solution of MICP problems using the Julia programming language and the
JuMP modeling language for optimization. In particular, we show how
mixed integer optimal control problems where the variables are
polynomials can be easily modeled and solved by seamlessly combining
several Julia packages and JuMP extensions with the Julia-written MICP
solver Pajarito.jl. Finally, we introduce the Julia-based interior
point solver for general non-symmetric cones Hypatia.jl, and show how
its use of non-standard cones can allow it to outperform commercial
solvers in certain instances.

Bio

Juan Pablo Vielma is the Richard S. Leghorn (1939) Career Development
Associate Professor at MIT Sloan School of Management and is
affiliated to MIT’s Operations Research Center. Dr. Vielma has a B.S.
in Mathematical Engineering from University of Chile and a Ph.D. in
Industrial Engineering from the Georgia Institute of Technology. His
current research interests include the theory and practice of
mixed-integer mathematical optimization and applications in energy,
natural resource management, marketing and statistics. In January of
2017 he was named by President Obama as one of the recipients of the
Presidential Early Career Award for Scientists and Engineers (PECASE).
Some of his other recognitions include the NSF CAREER Award and the
INFORMS Computing Society Prize. He is currently an associate editor
for Operations Research and Operations Research Letters, a member of
the board of directors of the INFORMS Computing Society, and a member
of the NumFocus steering committee for JuMP.

We strongly encourage students to act now to seek full-time employment, internships, and projects (rather than waiting until the end of the semester). Plus, enjoy a free pizza lunch!

EVENT DETAILS

Where: ISyE Main Bldg, 2nd Floor Atrium

When: Wednesday, September 11, 2019 | 11am - 2pm

What: The session will include:

• Networking opportunities in the ISyE atrium
• Food and refreshments

Please plan on staying for the duration of the event and bring copies of your resume and business cards. Dress is business casual.

REGISTER ONLINE by noon September 2nd to upload your resume and have it forwarded to the organization representatives!

EVENT SPONSOR

The event is sponsored through the generosity and support of JP Morgan Chase & Co. and APICS - Atlanta Chapter. APICS is a non-profit educational organization addressing operations management and supply chain management issues, and providing professional development opportunities to its members. APICS Membership is free for full time students. Join today at www.apics.org/join and start networking at local APIC Atlanta events. Also make sure to stop by the APICS table at the event.

Deep Generative Models in the Diffusion Limit

Abstract: 

In deep generative models, the latent variable is generated by a time-inhomogeneous Markov chain, where at each time step we pass the current state through a parametric nonlinear map, such as a feedforward neural net, and add a small independent Gaussian perturbation. In this talk, based on joint work with Belinda Tzen, I will discuss the diffusion limit of such models, where we increase the number of layers while sending the step size and the noise variance to zero. I will first provide a unified viewpoint on both sampling and variational inference in such generative models through the lens of stochastic control. Then I will show how we can quantify the expressiveness of diffusion-based generative models. Specifically, I will prove that one can efficiently sample from a wide class of terminal target distributions by choosing the drift of the latent diffusion from the class of multilayer feedforward neural nets, with the accuracy of sampling measured by the Kullback-Leibler divergence to the target distribution. Finally, I will briefly discuss a scheme for unbiased, finite-variance simulation in such models. This scheme can be implemented as a deep generative model with a random number of layers.

Bio: 

Maxim Raginsky received the B.S. and M.S. degrees in 2000 and the Ph.D. degree in 2002 from Northwestern University, all in Electrical Engineering. He has held research positions with Northwestern, the University of Illinois at Urbana-Champaign (where he was a Beckman Foundation Fellow from 2004 to 2007), and Duke University. In 2012, he has returned to the UIUC, where he is currently an Associate Professor with the Department of Electrical and Computer Engineering, the Coordinated Science Laboratory, and the Department of Computer Science. His research interests cover probability and stochastic processes, deterministic and stochastic control, machine learning, optimization, and information theory. Much of his recent research is motivated by fundamental questions in modeling, learning, and simulation of nonlinear dynamical systems, with applications to advanced electronics, autonomy, and artificial intelligence.

SESSION OVERVIEW

The Hunter River region on Australia’s east coast is home to the world’s largest coal export operation. More tons of coal are exported each year from the port of Newcastle, at the mouth of the Hunter River, than from any other port in the world. Coal exported is mined by more than 10 competing producers operating upward of 40 mines, but uses shared infrastructure to reach the port, be blended into final products held in stockyards and loaded onto vessels at berth. The shared infrastructure, which consists of rail lines, trains, stockyards, stockpile stacking and reclaiming machinery, berths, vessel loading machinery, the river channel and tug boats, is owned and/or operated by several different organizations. All of these organizations must coordinate their activities so as to deliver coal in an effective and efficient way. In this talk, we will discuss how collaboration between the many stakeholders has been shaped and evolved, how collaboration has been essential to the delivery of productivity gains in this complex logistics system, and will touch on the critical role analytics and the tools of operations research have played.​

SESSION SPEAKER

Natashia Boland, Fouts Family Professor, Stewart School of Industrial and Systems Engineering

Register Online for upcoming SCL IRC seminars

Attendance to the sessions is complimentary for SCL corporate partners, SCL Industry Advisory Board members, SCL affiliated faculty and students, and students enrolled in the Masters in Supply Chain Engineering program.

To take advantage of the included lunch, you must register by deadline noted on the registration page.

If you have any questions, please email event@scl.gatech.edu.

• The Project Management Institute is the world's professional society for project management, maintains robust body of knowledge and certification programs
• Atlanta is home to the world's 3rd highest number of certified project managers
• PMI was founded 50 years ago, at Georgia Tech, by Georgia Tech alum Jim Snyder, with assistance by SCL's predecessor organization
• Project Management methods are crucial to the success of engineers, consultants, analysts and a key differentiator in job search and career progression

On September 9, a number of events associated with the 50th anniversary will take place on campus. Including SCL's program:

What: Project Management Profession – its Roots, Growth, and Future

• The World of Project Oriented Work – From 1960 to 2019
• History of PMI and Georgia Tech's Role
• A Celebration of 50 Years
• The Changing World of Managing Projects

Speaker: PMI founder and Tech Alum Jim Snyder

Mr. Snyder is a founder of the Project Management Institute, a Fellow of the Institute, and a past Volunteer Executive Director, President and Chairman of the Board. He is Chair Emeritus of the Board of Directors of the PMI Educational Foundation, and Advisor to the Board of the Delaware Valley Chapter. Mr. Snyder, who is casually referred to as the "Father of PMI," has been awarded the PMI Man of the Year Award and is an Honorary Life Member of the Institute. He is also a founding member of the PMI Delaware Valley Chapter and the Pharmaceuticals Specific Interest Group (SIG). Mr. Snyder served as the project manager for the construction of PMI's World Headquarters in Newtown Square, Pennsylvania, USA. The PMI James R. Snyder Center for Project Management Knowledge and Wisdom, located in the headquarters building, is named in his honor.

Where: Georgia Tech Manufacturing Institute Auditorium (813 Ferst Drive, N.W., Atlanta, GA 30332-0560)

When: Monday, September 9, Noon to 1:30 (Pizza and Drinks provided)

Register Online to Attend

Attendance to the sessions is complimentary for SCL corporate partners, SCL Industry Advisory Board members, SCL affiliated faculty and students, and students enrolled in the Masters in Supply Chain Engineering program.For the general public, there is a $12 meal fee.

To take advantage of the included meal, you must register by deadline noted on the registration page.

If you have any questions, please email event@scl.gatech.edu.

Timing it Right: Balancing Inpatient Congestion versus Readmission Risk at Discharge

Abstract:

When to discharge a patient plays an important role in hospital patient flow management and patient outcomes. In this work, we develop and implement a practical decision support tool to aid hospitals in managing the delicate balance between readmission risk at discharge and ward congestion. We formulate the discharge decision framework as a large-scale Markov Decision Process (MDP) that integrates a personalized readmission prediction model to dynamically prescribe both how many and which patients to discharge each day. We overcome challenges from both the analytical and prediction sides. Due to patient heterogeneity and the fact that length-of-stay is not memoryless, the MDP suffers the curse of dimensionality. We derive useful structural properties and leverage an analytical solution for a special cost setting to transform the MDP into a univariate optimization; this leads to an efficient dynamic heuristic. Meanwhile, off-the-shelf prediction models alone could not provide adequate input for our decision support framework. To bridge this gap, we integrate several statistical methods to build a new readmission prediction model that allows us to implement our decision framework with existing hospital data systems.

Through extensive counterfactual analyses, we demonstrate the value of our recommended discharge policy over our partner hospital’s historical discharge behavior. We also discuss the implementation efforts of this discharge optimization tool at our partner hospital. This is joint work with Jonathan Helm at Kelley School of Business at Indiana University and industry partners. The paper received the Pierskalla Best Paper Award at INFORMS 2018 and can be accessed at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3202975

Bio:

Pengyi Shi joined the Krannert School of Management at Purdue University as an Assistant Professor in January 2014. She received her Ph.D. degree in Industrial Engineering from Georgia Institute of Technology before joining Purdue. Her research interests include healthcare operations and stochastic modeling, with a focus on hospital patient flow management. She has collaborated with practitioners and faculty members from different healthcare organizations, including major hospitals in the US, Singapore, and China.

Abstract:

AI-based systems, using Machine Learning models, are powering more and more processes in our daily lives, analyzing data to recognize patterns of behavior, and making recommendations for everything from what movies to binge next to whether your loan application is approved. They are being adopted to organize global supply chains, and provide the foundation for military weapon systems. What are the risks of using this technology in ever expanding ways? What are the privacy and ethical challenges when we have only the biased data from human history on which to train them? How do we make systems that interface with individual humans in ways that don't create an unmanageable imbalance of power? Overcoming the obstacles to achieve fair and just systems that strengthen rather than fray our social structures will not be easy. We must consider these questions in the light of human rights and available governance strategies, with a clear commitment to the ideals that have enabled the growth of human freedoms.

Bio:

In Ms. Brenda Leong's current appointment as the Senior Counsel & Director of Strategy with the Future of Privacy Forum, she oversees strategic planning of organizational goals, as well as managing the FPF portfolio on biometrics, particularly facial recognition, along with the ethics and privacy issues associated with artificial intelligence. She works on industry standards and collaboration on privacy concerns, by partnering with stakeholders and advocates to reach practical solutions to the privacy challenges for consumer and commercial data uses.

Ms. Leong has recently had several Future of Privacy Forum publications and has been featured in digital publications including the Bulletin of Atomic Scientists. This past January, Leong along with Evan Selinger presented their paper “Robot Eyes Wide Shut, Understanding Dishonest Anthropomorphism,” during the ACM FAT* Conference 2019.

Ms. Brenda Leong received her Master of Science in International Relations from Troy State University and her Juris Doctor from the George Mason School of Law at George Mason University.

Abstract:

The Centers for Medicare and Medicaid Services (CMS) Star Ratings methodology for publicly evaluating hospitals uses a latent variable model that is based on the presumption of a single, but unobservable, hospital-specific quality factor shared across a group of performance measures. Performance measures are given higher weight if they statistically appear to be more strongly correlated with this hidden factor. We show how this approach, when applied to measures that are weakly or not correlated with each other, can effectively ignore measures and can exhibit "knife-edge" instability, so that even if hospitals improve relative to all other hospitals, they may nonetheless score lower overall due to weight shifting onto different measures than before.

In contrast, we provide an approach to scoring and ranking hospitals that, under reasonable conditions, ensures hospitals that improve relative to all other hospitals obtain higher scores, while also having the capability to autonomously adjust weights as measures are added or subtracted over time. Rather than exploit statistical correlation, we propose a conic optimization framework that offers a new integrated approach in Data Envelopment Analysis (DEA) for simultaneous efficiency analysis and performance evaluation. We develop theory that explains the behavior of our approach, including various properties satisfied by hospital scores at optimality. Using data we apply our approach to score and rank nearly every hospital in the United States, and demonstrate the extent to which it agrees or disagrees with the existing approach to the CMS Star Ratings.

Bio:

Daniel Adelman, PhD., is the Charles I. Clough Jr. Professor of Operations Management at the University of Chicago Booth School of Business.   He leads the Healthcare Analytics Laboratory at Chicago Booth, in which teams of graduate students and post-doctoral researchers work on real-world projects with providers to improve healthcare delivery through the analysis of large datasets.   Adelman also developed and teaches a course on Healthcare Data Analytics that draws students from around the University of Chicago campus.  He has published research based on work with various industries, including the electricity smart grid, gasoline supply chains, software-release planning, intermodal logistics, manufacturing, revenue management/pricing, and healthcare delivery. He publishes regularly in leading academic journals, and he is Area Editor for the Operations and Supply Chains Area of Operations Research. He teaches in Chicago Booth's Executive MBA Program, and in Booth’s Executive Program for Emerging Leaders.  He serves on the faculty advisory board of the Harry L. Davis Center for Leadership. Adelman received a PhD in Industrial Engineering and Operations Research in 1997 from the Georgia Institute of Technology, where he also received a bachelor's degree in industrial engineering and a Master of Science in Operations Research.   Adelman joined the Chicago Booth faculty in 1997.

Abstract:

Electric power systems are critical infrastructure that underlie almost all aspects of modern society. With rapidly increasing quantities of renewable generation and the continuing expansion of electricity markets, electric power systems are undergoing significant changes. New algorithms for optimizing the design and operation of electric power systems are needed in order to enable these transformational changes.

This presentation focuses on the "power flow equations" which model the physical relationships that exist on electric transmission and distribution systems. The nonlinearity of the power flow equations results in a variety of algorithmic and theoretical challenges, including non-convex feasible spaces for optimization problems constrained by these equations. Many convex relaxation techniques have recently been applied to simplify the power flow representations used in a variety of power system optimization and control problems. This presentation first overviews various convex relaxation techniques that have been applied to the power flow equations, with a focus on semidefinite programming and second-order cone programming formulations as well as techniques for tightening the relaxations. This presentation then summarizes several applications of convex relaxations in a variety of power systems contexts.

Bio:

Daniel Molzahn is an assistant professor in the School of Electrical and Computer Engineering at the Georgia Institute of Technology. Daniel also holds an appointment as a computational engineer in the Energy Systems Division at Argonne National Laboratory. Daniel was a Dow postdoctoral fellow at the University of Michigan. He completed the B.S., M.S., and Ph.D. degrees in Electrical Engineering and the Masters of Public Affairs degree from the University of Wisconsin–Madison, where he was a National Science Foundation Graduate Research Fellow. His research interests are in the application of optimization techniques to electric power systems.

Amazon is currently hiring around the nation, in a variety of different operations roles. Join us to learn about Amazon *opportunities for Georgia Tech students!
 

KICK OFF DAY TABLING

Monday, August 26, 2019| 10am - 2pm
Tech Walkway, southwest corner (tent and tables) weather permitting OR ISyE Atrium if we are rained out.

Stop by to chat with Amazon representatives and learn about the opportunities they are currently hiring for.

GENERAL INFORMATION SESSION / PANEL DISCUSSION

Tuesday, August 27, 2019 | 12pm - 2pm
ISyE 2nd floor Atrium (Main Building, 755 Ferst Drive)

Learn about Amazon Operation's culture and how your background can contribute to our growing network.

AMAZON OFFICE HOURS ON CAMPUS

Tuesday, August 27 | 2–5 PM | ISyE Rooms 226 & 202 (Groseclose/765 Ferst)

Wednesday and Thursday, August 28 & 29 | 8 AM – 5 PM | ISyE Rooms 226 & 202 (Groseclose/765 Ferst)

Meet Amazon team members, learn more about the culture, and ask any questions...

*If interested in attending a related event and talking with the Amazon team, please register and upload your resume at http://bit.ly/amazongt2019.

Do We Necessarily Have More Cost-Effective Information with Bigger Data?

Abstract

It is a normally held belief that more data provide more information. Such a feeling is widespread under big data movement, but largely without mentioning what kind of information one may need or be willing to utilize. In order to hold a reasonable discourse, I begin by exploring under certain stochastic models of one kind or another and then try to grasp what additional useful information may or may not entail as more data came in.

I will illustrate situations where a sentiment that more data may amount to more useful information is nearly valid. But, I will also show that the same belief may not be entirely justified across the board. That is, any relative gain in information may not be substantial as one utilizes more available data. Indeed, the relative gain in information may go down with more data.

In this presentation, I will share some preliminary ideas and analysis to highlight my thoughts. It is my earnest belief that there must be more to it than just using increased amount of available data simply because bigger datasets are out there!   

Biography

Prof. Nitis Mukhopadhyay received PhD degree (1975) from Indian Statistical Institute-Calcutta. He has been a full professor in the Department of Statistics, University of Connecticut-Storrs, USA since 1985. He served as the Head of this department from 1987-1990.

Prior to joining the University of Connecticut, Prof. Mukhopadhyay was a faculty member at the following institutions: Monash University-Melbourne, Australia (1976-77), University of Minnesota-Minneapolis (1977-78), University of Missouri-Columbia (1978-79), and Oklahoma State University-Stillwater (1979-85).

He made prolific contributions in a number of areas including statistical inference-parametric and nonparametric, sequential analysis, multiple comparisons, clinical trials, applied probability, econometrics, and applications. Prof. Mukhopadhyay is especially recognized for path-breaking contributions in (i) sequential analysis as well as (ii) selection and ranking.  His honors include elected Fellows of the Institute of Mathematical Statistics (2002), the American Statistical Association (2003), the American Association for the Advancement of Science (2012), elected Member of the International Statistical Institute (2007), elected Member of Connecticut Academy of Arts and Sciences (2014), Fellow of the Royal Statistical Society, the Abraham Wald Prize in Sequential Analysis (2008), the Don Owen Award (2015).

He was awarded the Honorary Fellowship by the Institute of Applied Statistics Sri Lanka (2017). He has been the Editor-in-Chief for the premier journal, Sequential Analysis, since 2004 and serves as an Associate Editor for a number of other leading international journals. 

As an author or co-author, Prof. Mukhopadhyay published 6 books, 18 book chapters, nearly 300 peer-reviewed research papers, and edited or co-edited more than 7 special volumes. He has supervised 26 Ph.D. students as a major adviser and has 4 Ph.D. continuing under his guidance at this time. His former PhD advisees have gone to academia, businesses, and industries to make their own marks.

SESSION OVERVIEW

There are numerous opportunities in utilizing quantitative methods in health and humanitarian systems. There are often limited resources and multiple stakeholders in these settings, and efficient and effective decision-making is important for resource utilization as well as improved outcomes for individuals or populations. We will discuss a few examples of the applications of quantitative methods in health and humanitarian systems, such as pre-positioning inventory for disaster preparedness, post-disaster debris management, intervention strategies and resource allocation for disease management, and allocation of healthcare workers to geographic areas.

SESSION SPEAKER

Pinar Keskinocak, William W. George Chair and Professor, ISyE and ADVANCE Professor, College of Engineering and and Director of the Center for Health and Humanitarian Systems

Register Online for upcoming SCL IRC seminars

Attendance to the sessions is complimentary for SCL corporate partners, SCL Industry Advisory Board members, SCL affiliated faculty and students, and students enrolled in the Masters in Supply Chain Engineering program.

To take advantage of the included lunch, you must register by deadline noted on the registration page.

If you have any questions, please email event@scl.gatech.edu.

Supply Chain Management projects can span a wide range of project types including supply chain strategy, network analysis, facility design build, supply chain technology selection and implementation, and continuous process improvement initiatives. This course provides an overview of project management methodologies as applied in the supply chain environment. Class discussion and projects provide an understanding of how fundamental project management approaches and industry best practices can be used to effectively manage the complexities. Supply chain projects typically require managing resources, stakeholder alignment, risk management, customer impact, and effective communication across many internal and external business partners.

HOW YOU WILL BENEFIT

Upon completion of this course, you will be able to:

• Identify the business processes and functional organizations impacted by specific types of supply chain projects
• Understand the leading project methodologies (ITIL, CSM, etc.) and why complex supply chain projects require elements from each to be successful
• Understand the components of Project Management Book of Knowledge and how they can be applied in the supply chain environment
• Define project requirements and expectations for supply chain projects
• Develop the Work Breakdown Structure (WBS) for supply chain projects
• Identify the critical path for a project from its network diagram
• Be able to form and deploy a supply chain project team
• Understand how to assess, manage, and mitigate project risk
• Properly monitor, control, and close the project

WHAT IS COVERED

• Overview of Leading Project Management Methodologies & Certifications
• Types of Supply Chain Projects – Process/Functional Impacts
• PMBOK knowledge areas in context of Supply Chain Projects
• Overview of PMBOK knowledge areas in context of Supply Chain Projects
• Overview of Business Process Analysis and the SCOR model
• Defining the Project & the Effective Supply Chain Project Manager
• Why Supply Chain Projects Fail
• Defining Supply Chain Project Requirements, Teams, Roles & Responsibilities
• Traditional Phases of a Supply Chain Project
• Developing the Work Breakdown Structure for SC Projects
• Review of the Supply Chain Work Breakdown Structure
• The use of Software to Support Project Management
• Estimating Cost, Duration, and Resources
• Resource Management & Developing Dependency Diagrams
• Defining, Assessing, & Mitigating Risk
• Monitoring & Controlling the Project Plan
• Closing the Project

ON-CAMPUS COURSE MATERIALS

• Notebook of slides, notes, exercises and project management tools and templates

COURSE PREREQUISITES

None.

CERTIFICATE INFORMATION

For those interested in earning the Supply Chain Project Management Certificate, this course is the first of the four-course certificate program. To earn the certificate, participants must register and complete the following courses in the below sequence within four years, plus one elective.

1. Supply Chain Project Management: Fundamentals
2. Supply Chain Project Management: Vendor Selection & Management
3. Supply Chain Project Management: Effectively Managing Transformation Projects

For a list of courses that can be used as electives towards this certificate, please visit the Georgia Tech Professional Education website.

This accelerated learning boot camp covers 2.5 days of instructor lead facilitation and is designed to cover key learning points of the CSCP learning system. The learning system content is included in the course. This boot camp is beneficial for the self-study learner planning to take the APICS CSCP certification exam.

The APICS Certified Supply Chain Professional™ (CSCP) program takes a broad view of the field, extending beyond internal operations to encompass all the steps throughout the supply chain--from the supplier, through the company, to the end consumer--and how to effectively manage the integration of these activities to maximize the company’s value chain.

The CSCP program provides candidates with the necessary tools to effectively manage global supply chain activities and enables them to implement best practice approaches to increase supply chain efficiencies.

WHO SHOULD ATTEND

The APICS CSCP is designed for supply chain professionals and people managing and planning extended supply chains especially people in:

• Supply chain design
• Supplier management
• Transportation
• Supply chain management
• Distribution channels
• 3PL, 4PL management
• Customer management
• Supply chain consulting

HOW YOU WILL BENEFIT

• Master advanced supply chain management principles that extend beyond an organization’s internal end-to-end operations from suppliers to customers
• Apply supply chain knowledge and analytical skills to streamline operations and produce bottom line results
• Explain the role of each element of the integrated supply chain concept
• Understand how successful supply chain management adds value to your organization
• Learn how to develop a supply chain strategy that aligns with corporate strategy
• Understand natural dynamics within the supply chain to optimize performance and profitability
• Evaluate the process constraints and choices within Global Logistics to establish a plan linked to overall strategy
• Effectively use customer data to improve service performance and increase value to suppliers and customers
• Understand the strategic importance of purchasing and supply relationships
• Understand the innovative technologies enabling collaborative commerce and global visibility
• Apply technology to enhance performance of distribution, reverse logistics, and global supply chain communications

WHAT IS COVERED

Supply Chain Design

• Addresses the concepts and strategies used for developing a supply chain strategy aligning with business goals and corporate strategy
• Designing a supply chain for the flow of product, information and cash
• Understanding and using the SCORreg model

Supply Chain Planning and Execution

• Addresses the processes required to procure and deliver goods and services
• Management of demand and supply relationships

Supply Chain Planning and Execution

• Related concepts of order management and customer service
• Defining and measuring logistics customer service outputs

Supply Chain Improvement and Best Practices

• Managing and balancing supply and demand through measuring, analyzing and improving supply chain processes
• Compliance with standards and regulations
• The importance of sustainable best practices and social responsibility
• Assessment of risk within the supply chain

COURSE MATERIALS

Participants will be provided the APICS CSCP course reading material and one-year access to the APICS online courseware.

COURSE PREREQUISITES

None.

PROGRAM TIMES

On the first day, please check in at least 30 minutes before the class start time.

• First Day - 1pm to 5:00pm (lunch not included)
• ​Second Day - 8:00am to 5:00pm
• Third Day - 8:00am to 5:00pm

This accelerated learning boot camp covers 2.5 days of instructor lead facilitation and is designed to cover key learning points of the CLTD learning system. The learning system content is included in the course. This boot camp is intended for the self-study learner planning to take the APICS CLTD certification exam.

The APICS Certified in Logistics, Transportation and Distribution™ (CLTD) certification provides you with the information you need to meet today’s supply chain logistics challenges.

Created by APICS and a team of subject matter experts, the CLTD designation covers a comprehensive body of knowledge that sets the global standard for best practices in logistics, transportation and distribution.

WHO SHOULD ATTEND

The APICS CLTD is designed for logistics professionals and people planning and managing global logistics across all industries, especially:

• Logistics managers/engineers
• Supply chain managers
• Traffic managers
• Transportation/fleet managers
• Warehouse operations/distribution managers
• Reverse logistics managers
• Supply chain/operations consultants

No matter where you are in your career, earning the CLTD credential will demonstrate your mastery of a broad range of logistic, transportation and distribution topics.

HOW YOU WILL BENEFIT

• Master the essential knowledge needed for the logistics, transportation and distribution industry
• Bring new ideas to your organization
• Remain current with global logistics trends and developments
• Boost your confidence with recognition as a logistics expert
• Reduce costs and impact your organization’s bottom line
• Allow you to more effectively communicate with others across your extended supply chain

Participants will gain an understanding of the processes, best practices, and tools used by today’s logistics managers and how to design and administer systems to control the movement and positioning of material and information flows to satisfy customer requirements at the lowest total cost.

WHAT IS COVERED

Logistics and Supply Chain Overview

• Overview of macro and micro logistics processes and systems defining material and information flows
• Achieving consistent customer service at the lowest total cost
• Illustrate logistics cross-functional relationships with marketing, finance and manufacturing

Capacity Planning and Demand Management

• Understand strategic and operational considerations, and important trade-offs related to logistics, procurement, manufacturing and customer relationship management
• Developing resource-alignment and organizational forecasts

Order Management

• Related concepts of order management and customer service
• Defining and measuring logistics customer service outputs

Inventory and Warehouse Management

• Review methods and techniques firms use to control, manage, and store inventory
• Understand economic and service benefits warehousing delivers
• Define warehouse types and operating characteristics

Transportation

• Understand major principles and processes of transportation management
• Review fundamental mode roles, capabilities and responsibilities

Global Logistics Considerations

• Navigating today’s more complex supply chains from a total cost perspective
• Using transportation modal analysis, security considerations and regulations

Logistics Network Design

• Understand how transportation and inventory economics impact network design considerations
• Introduction to modeling tools and techniques
• Working with the processes, resources and tools for managing risk

Reverse Logistics Considerations

• A look at the expanded view of reverse supply chain flows and sustainability
• What is logistics social responsibility and the triple bottom line

COURSE MATERIALS

Participants will be provided the APICS CLTD course reading material and one-year access to the APICS online courseware.

COURSE PREREQUISITES

None.

PROGRAM TIMES

On the first day, please check in at least 30 minutes before the class start time.

• First Day - 8:00am to 5:00pm
• ​Second Day - 8:00am to 5:00pm
• Third Day - 8:00am to 12:00pm (lunch not included)

Abstract:

We consider modeling of coupled systems using an equilibrium framework.The systems are coupled by market constraints and can be effectively modeled in the EMP framework.Standard and new mechanisms to model resulting complementarity constraints will be outlined, and the agents problems will be formulated to allow risk preferences. Some computational results related to energy systems and other applications will be given.

Bio:

Dr. Ferris' research is concerned with algorithmic and interface development for large scale problems in mathematical programming, including links to the GAMS and AMPL modeling languages, and general purpose software such as PATH, NLPEC and EMP. He has worked on many applications of both optimization and complementarity, including cancer treatment planning, energy modeling, economic policy, traffic and environmental engineering, video-on-demand data delivery, structural and mechanical engineering.

Ferris is a SIAM fellow, an INFORMS fellow, received the Beale-Orchard-Hays prize from the Mathematical Programming Society and is a past recipient of a NSF Presidential Young Investigator Award, and a Guggenheim Fellowship. He serves on the editorial boards of Mathematical Programming, and Optimization Methods and Software.

Parking: Free after 4pm (garage entrance is on Harrison, between Terry and Boren on the north side of the building).

Food & Drink: GT will be providing the room and snacks; a cash bar and additional food are available for purchase.

Please RSVP via Evite to let us know you are coming!
 

Go Jackets!

Advances and Challenges in Conformal Inference

Abstract:

We describe some recent advances in distribution-free prediction intervals in regression, using the conformal inference framework. This framework allows for the construction of a prediction band for the response variable using any estimator of the regression function. The resulting prediction band preserves the consistency properties of the original estimator under standard assumptions, while guaranteeing finite-sample marginal coverage even when these assumptions do not hold. We discuss two major variants of the conformal framework: full conformal inference and split conformal inference, along with a related jackknife method. These methods offer different tradeoffs between statistical accuracy (length of resulting prediction intervals) and computational efficiency.  Time permitting, we will discuss extensions and some open challenges related to conformal inference.

This represents joint work with Rina Barber, Emmanuel Candes, Max G’Sell, Jing Lei, Aaditya Ramdas, Alessandro Rinaldo, and Larry Wasserman.

Bio:

Ryan Tibshirani is an Associate Professor jointly appointed in the Departments of Statistics and Machine Learning at Carnegie Mellon University.  He is the recipient of an NSF CAREER Award in 2016, and a Carnegie Mellon Teaching Innovation Award in 2017.  He joined the Statistics faculty at Carnegie Mellon University in 2011, and joined the Machine Learning faculty in 2013. He did his Ph.D. in Statistics at Stanford University in 2011 (under Jonathan Taylor), and his B.S. in Mathematics at Stanford University in 2007.

Sequential Selection for Accelerated Life Testing via Approximate Bayesian Inference

Abstract: 

Approximate Bayesian inference (Chen and Ryzhov, 2019) has been proposed to construct computationally tractable statistical learning procedures for incomplete or censored data. In this talk, I will discuss a sequential model-updating procedure via approximate Bayesian inference for the Log-normal model with censored observations. We show that the proposed procedure leads to a consistent model parameter estimation. The developed model updating procedure also enables a closed form expression of a sequential design criterion. The proposed procedure is applied to accelerated life testing experiments, which aims at determining the material alternative with the best reliability performance. 

Bio:

Dr. Qiong Zhang is an assistant professor in the School of Mathematical and Statistical Sciences at Clemson University. Previously, she was an assistant professor of statistics at Virginia Commonwealth University in 2014–2018. Dr. Zhang received a B.S. degree in statistics from Nankai University and an M.S. degree in statistics from Peking University in 2007 and 2009, respectively. She received her Ph.D. degree in statistics from University of Wisconsin-Madison in 2014. Dr. Zhang’s research interests include the interface between information collection and statistical modeling, design and analysis of computer experiment, and uncertainty quantification.

Bid Package Development and Supplier Selection is designed to deepen participants' knowledge base of core activities in the procurement & supply management function. The program covers the sourcing process, specifications gathering, common bid package alternatives, cross-functional collaboration and supplier evaluation & selection. Participants will walk away ready to develop bid packages more thoroughly to help drive sourcing decisions for their organizations. This "hands on" delivery focuses on the professional serving as the main liaison between the buying organization and the selling organization in the company sourcing process.

Who Should Attend

This course is ideal for sourcing initiative leaders, procurement professionals, project managers, finance analyst, contract managers and all procurement & supply management-related professionals involved with bid package development, bid package analysis, negotiations preparation, contracting and supplier selection activity.

How You Will Benefit

• Increase knowledge base of solicitation options (i.e. RFI, RFP, RFQ) and which solicitation approach to apply based on your organization's needs
• Enhance understanding of sourcing process and critical steps in the Bid Package development and delivery activities
• Better leverage and utilization of solicitation tools to drive successful development of bid packages
• Improve set up and execution of supplier selection scorecards to aid in identifying best Total Cost of Ownership alternatives
• Heighten understanding of executive communication to leverage leadership support throughout the organization

What Is Covered

• Bid Package Development Overview
• Sourcing Initiative Process Roadmap 
• Marketplace Analysis Tools
• Bid Package Alternatives & Design
• Supplier Selection & Communication

Essential Principles of Negotiations level-sets the participants' understanding of negotiation influence and strengthens preparation, planning and execution activities involved with both simple and complex negotiations. The program includes industry techniques and tools for traditional supplier negotiations, as well as tips for internal cross-functional leadership. Participants walk away with a standard industry and customized individual experience which includes their personal Negotiation Style “DNA” to help them embrace their own natural tendencies and strengths. The program includes mock negotiations to reinforce techniques and tactics immediately in a “no judgement zone” environment.

Who Should Attend

This course is ideal for sourcing initiative leaders, project leaders, business unit leaders, operations managers, sales leaders and procurement & supply management-related professionals who are involved with supplier selection, contract development and supplier performance management.

How You Will Benefit

• Increase emphasis negotiation conditioning and philosophy setting before and throughout the entire sourcing engagement process
• Enhance your toolbox of industry standard negotiation prep tools like the SWOT and BATNA
• Better prepare for negotiations by leveraging knowledge of key negotiation terms and counter-offer tactics
• Improve negotiation table techniques and soft skills to direct and redirect negotiation momentum
• Heighten ability to successfully utilize your traditional "comfort zone" approach in combination with your negotiation team’s strengths by leveraging Personal Negotiation Styles

What Is Covered

• Negotiation Conditioning Overview
• Negotiation Preparation Tools
• Negotiation Execution Techniques
• Stakeholder Engagement & Team Leadership
• Live Negotiations Simulation & Feedback

Contracting Excellence provides participants with a holistic and integrated understanding of contract law, contract types, key industry standard contract terms, and contract structure to improve their confidence when creating or modifying contract documents. The program is geared to reinforce standards of excellence for professionals who are responsible for delivering contractual agreements and mitigating financial risk for their organization.

Who Should Attend

This course is ideal for contract managers, procurement professionals, sourcing initiative leaders, project managers and all procurement & supply management-related professionals involved with bid contract development, contract execution or supplier performance management.

How You Will Benefit

• Increase knowledge base in law of agency concepts to gain more confidence with contracting.
• Enhance understanding of industry standard contract types and contract terms for more robust contract agreements.
• Better leverage sourcing category knowledge to modify existing contract elements for more holistic contract agreements. 
• Improve internal contract execution communication for better results.
• Heighten sense of executive financial impact and risk needs to gain leadership early support.

What Is Covered

• Law of Agency Overview
• Contract Agreement Types
• Defining Key Terms
• Contract Structure & Drafting
• Risk Mitigation & Communication

EzGP: Easy-to-Interpret Gaussian Process Models for Computer Experiments with Both Quantitative and Qualitative Factors.

Abstract:

Computer experiments with both quantitative and qualitative (QQ) inputs are commonly used in science and engineering applications. Constructing desirable emulators for such computer experiments remains a challenging problem. In this article, we propose an easy-to-interpret Gaussian process (EzGP) model for computer experiments to reflect the change of the computer model under different level combinations of qualitative factors. The proposed modeling strategy, based on an additive Gaussian process (GP), is flexible to address the heterogeneity of computer models involving multiple  qualitative factors. We also develop two useful variants of the EzGP model to achieve computation efficiency when dealing with high dimensional data and large data size. The merits of these models are illustrated by a real data application and several numerical examples.

Bio:

Dr. Qian Xiao is currently an assistant professor in statistics at University of Georgia. He received his Ph.D. at UCLA in 2017. His research interests are experimental design, stochastic modeling and pharmaceutical studies.

Bridging convex and nonconvex optimization in noisy matrix completion: Stability and uncertainty quantification 

Abstract:  

This talk is concerned with noisy matrix completion: given partial and corrupted entries of a large low-rank matrix, how to estimate and infer the underlying matrix? Arguably one of the most popular paradigms to tackle this problem is convex relaxation, which achieves remarkable efficacy in practice. However, the statistical stability guarantees of this approach is still far from optimal in the noisy setting, falling short of explaining the empirical success. Moreover, it is generally very challenging to pin down the distributions of the convex solution, which presents a major roadblock in assessing the uncertainty, or “confidence”, for the obtained estimates --- a crucial task at the core of statistical inference.  

Our recent work makes progress towards understanding stability and uncertainty quantification for matrix completion. (1) We demonstrate that the convex programming approach achieves near-optimal estimation errors vis-a-vis random noise; (2) we develop a de-biased estimator that admits entrywise distributional characterizations, thus enabling asymptotically optimal inference.  All of this is enabled by bridging convex relaxation with the nonconvex Burer-Monteiro approach, a seemingly distinct algorithmic paradigm that is provably robust against noise. 

This is based on joint work with Cong Ma, Yuling Yan, Yuejie Chi, and Jianqing Fan. 

Bio:

Yuxin Chen is currently an assistant professor in the Department of Electrical Engineering at Princeton University. Prior to joining Princeton, he was a postdoctoral scholar in the Department of Statistics at Stanford University, and he completed his Ph.D. in Electrical Engineering at Stanford University. His research interests include high-dimensional statistics, convex and nonconvex optimization, statistical learning, and information theory. He received the 2019 AFOSR Young Investigator Award.

Kaitlyn Banks with Atlanta CSCMP will give an overview of the organization and discuss:

  • The influence of CSCMP for current undergraduate and graduate students
  • The power of networking with CSCMP
  • The many informative educational events CSCMP hosts

ABOUT CSCMP

The Council of Supply Chain Management Professionals (CSCMP) is the leading global association for supply chain management professionals and have been helping its members and their companies since 1963. Come learn more about how you can benefit as a student by joining CSCMP.

Low Rank Independence Samplers in Hierarchical Bayesian Inverse Problems
Andrew Brown
School of Mathematical and Statistical Sciences
Clemson University

Abstract:

In Bayesian inverse problems, the posterior distribution is used to quantify uncertainty about the reconstructed solution. In fully Bayesian approaches in which prior parameters are assigned hyperpriors, Markov chain Monte Carlo (MCMC) algorithms often are used to approximate samples from the posterior. However, implementations of such algorithms can be computationally expensive. In this talk, I will present a computationally efficient scheme for sampling high-dimensional Gaussian distributions in ill-posed Bayesian linear inverse problems. The approach uses Metropolis-Hastings independence sampling with a proposal distribution based on a low-rank approximation of the priorpreconditioned
Hessian. I will present results obtained when using the proposed approach with
Metropolis-Hastings-within-Gibbs sampling in numerical experiments in image deblurring and computerized tomography. Time permitting, I will also briefly discuss applying the low-approximation idea in marginalization-based MCMC algorithms to improve the mixing behavior when compared to standard block Gibbs sampling.

Bio:

Andrew Brown earned his B.S. in Applied Mathematics from Georgia Tech in 2006. After briefly working for Porsche Cars North America in Atlanta, he went to the University of Georgia to earn his MS and PhD in Statistics under the direction of Nicole Lazar and Gauri Datta. He then joined the School of Mathematical and Statistical Sciences at Clemson University, where he is currently an Assistant Professor. He spent Spring of 2016 as a Visiting Research Fellow at the Statistical and Applied Mathematical Sciences Institute. His research interests include high-dimensional Bayesian modeling and computation, neuroimaging data analysis (particularly functional and structural MRI), computer
experiments, and uncertainty quantification. 

In this talk, we propose a methodology to predict descriptions of solutions to discrete stochastic optimization problems in very short computing time. We approximate the solutions based on supervised learning and the training dataset consists of a large number of deterministic problems that have been solved independently (and offline). Uncertainty regarding a subset of the inputs is addressed through sampling and aggregation methods. Our motivating application concerns booking decisions of intermodal containers on doublestack trains. Under perfect information, this is the so-called load planning problem and it can be formulated by means of integer linear programming. However, the formulation cannot be used for the application at hand because of the restricted computational budget and unknown container weights. The results show that standard deep learning algorithms allow to predict descriptions of solutions with high accuracy in very short time (milliseconds or less). A careful comparison with alternative stochastic programming approaches is provided.

Monitoring & Improving Quality via Consumer Complaints - A Statistical Engineering Case Study

Abstract:

Procter & Gamble (P&G) receives millions of consumer comments each year and requires an efficient data-mining algorithm to identify products with unexpectedly high complaint counts, as these suggest potential quality or safety issues (aka “signals”). This talk will present an overview of P&G’s adopted signal detection method, the Multi-Item Gamma Poisson Shrinker (MGPS), an empirical Bayesian method for disproportionality analysis. It will also discuss the application of text mining on consumer comments to find misclassified adverse events and identify fraudulent complaints.  I will finish the talk by relating certain aspects of this case study to the statistical engineering framework proposed by Roger Hoerl and Ron Snee in their 2017 American Statistician paper, “Statistical Engineering: An Idea Whose Time Has Come?”

Bio:

William Brenneman is a Research Fellow and Global Statistics Discipline Leader at Procter & Gamble in the Data and Modeling Sciences Department and an Adjunct Professor of Practice at Georgia Tech in the Stewart School of Industrial and Systems Engineering.  Since joining P&G, he has worked on a wide range of projects that deal with statistics applications in his areas of expertise: design and analysis of experiments, robust parameter design, reliability engineering, statistical process control, computer experiments, data science, machine learning and general statistical thinking.  He was also instrumental in the development of an in-house statistics curriculum.

He received a Ph.D. degree in Statistics from the University of Michigan, an MS in Mathematics from the University of Iowa and a BA in Mathematics and Secondary Education from Tabor College.  William is a Fellow of the American Statistical Association (ASA), a Fellow of the American Society for Quality (ASQ), and a member of the Institute of Mathematical Statistics and the Institute for Operations Research and Management Sciences.  He has served as ASQ Statistics Division Chair, ASA Quality and Productivity Section Chair and is currently serving as an Associate Editor for Technometrics. William also has seven years of experience as an educator at the high school and college level. 

In this talk, we focus on two estimation problems: i) sparse regression and ii) sparse and smooth signal recovery. The first one is known to be NP-hard; we show that the second one is equivalent to a submodular minimization problem and, hence, is polynomially solvable. For both problems, we derive a sequence of strong convex relaxations. These relaxations are based on the ideal (convex-hull) formulations for rank-one/pairwise quadratic terms with indicator variables. The new relaxations can be formulated as conic quadratic or semidefinite optimization problems in an extended space; they are stronger and more general than the state-of-the-art models with the reverse Huber penalty and the minimax concave penalty functions. Furthermore, the proposed rank-one strengthening can be interpreted as a non-separable, non-convex, unbiased sparsity-inducing regularizer, which dynamically adjusts its penalty according to the shape of the estimation error function without inducing bias for the sparse solutions. Computational experiments with benchmark datasets show that the new conic formulations are solved fast and result in near-optimal estimators for non-convex L0-problems. Moreover, the resulting estimators also outperform alternative convex approaches from a statistical perspective, achieving high prediction accuracy and good interpretability.

The talk is based on the following recent papers with Andres Gomez and Shaoning Han:

https://atamturk.ieor.berkeley.edu/pubs/rank-one.pdf

https://atamturk.ieor.berkeley.edu/pubs/signal-estimation.pdf

IPIC 2019 aims to provide an open forum for researchers, industry representatives, government officials and citizens to together explore, discuss, introduce leading edge concepts, methodologies, recent projects, technological advancements,start-up initiatives, for current and future Physical Internet implementation.

Conference topics include inter-connected logistics, PI fundamentals, business models, governance and implementation, cross-chain control, synchromodal transportation, IT systems, stakeholders and their roles. New business models, enabling technologies and experimentations already underway will be presented, making this meeting a unique opportunity to learn, network and discuss the latest results and challenges about interconnected logistics. And, because logistics is global, participants will be from all over the world including researchers, industrial and international institution members, local authorities and standardization committees.

Visit www.pi.events to learn more about the conference.

SCL IRC Seminar: An AI Framework Used in Crisis Management and Decision Making

Frederick Benaben, Visiting Professor | IMT Mines Albi - Industrial Engineering Research Center - FRANCE

Model-Driven Engineering (MDE) is meant, above other usages mainly in software design, to support the process of system design, especially the transitions and the coherency between models of different levels and different point of views. In the presentation, Dr. Benaben will introduce an original and innovative Artificial Intelligence (AI) framework dedicated to include MDE in AI technologies. An instantiation of this framework on the domain of crisis management will be presented as an illustrative example of providing sentience, agility and resilience in the handling of an instable situation. Then, the usage of this research framework will be extended to provide avenues and perspectives in ways to connect Industrial Engineering and Artificial Intelligence to support (collaborations of) organizations, decision making and management of organizations.

Register Online for upcoming SCLIRC seminars

Attendance to the sessions is complimentary for SCL corporate partners, SCL Industry Advisory Board members, SCL affiliated faculty and students, and students enrolled in the Masters in Supply Chain Engineering program.

To take advantage of the included lunch, you must register by deadline noted on the registration page.

If you have any questions, please email event@scl.gatech.edu.

Abstract:

Crowdsourcing systems often exhibit some negative externalities in the form of collecting and using sensitive data about their participants, such as their location. However, do workers care about these externalities, and do they take them into account when deciding to accept to reject a task? In this talk, we review the conflicting literature that tried to model the economic valuation of privacy. We then focus on mobile crowdsourcing as a test case that simplifies some of the experimental settings involved in privacy research. To assess the value users give to their location, and to understand its relationship with information sharing, we conducted a study on a mobile crowdsourcing platform. We quantify how users’ valuation of location privacy is dependent on the sharing scenario. For instance, when the location is to be shared with an untrusted advertiser, the users require a premium as compensation for their information. Additionally, benefit perception and trust are found to be connected with more frequent location sharing, while perceived risks and privacy concern are associated with sharing one’s location less frequently. We then discuss how valuation-based approaches can help us better understand privacy concerns and ground theoretical notions of privacy in concrete user behavior.

Short bio:

Eran is a visiting associate professor at Cornell Tech and a permanent member of the Department of Industrial Engineering at Tel-Aviv University. He is also the co-director of the IWiT (Interacting with Technoloy Lab). Eran’s research group is working on usable privacy and security, large-scale analysis of interactive behavior, and mobile computing. The research group is currently running several projects funded by agencies such as the Israeli Science Foundation (ISF), DARPA, European Union Horizon 2020 Program, and Israel Ministry of Science.

Auctions are widely used in practice. While also extensively studied in the literature, most of the developments rely on significant assumptions about common knowledge on the seller and buyers' sides. In this work, we study the fundamental problem of designing optimal prior-independent selling mechanisms. In particular, the seller faces buyers whose values are drawn from an unknown distribution, and only knows that the distribution belongs to a particular class. We analyze a competitive ratio objective, in which the seller attempts to optimize the worst-case fraction of revenues garnered compared to those of an oracle with knowledge of the distribution. Our results are along two dimensions. We first characterize the structure of optimal mechanisms. Leveraging such structure, we then establish tight lower and upper bounds on performance, leading to a crisp characterization of optimal performance for a spectrum of families of distributions. In particular, our results imply that a second price auction is an optimal mechanism when the seller only knows that the distribution of buyers has a monotone increasing hazard rate, and guarantees at least 71.53% of the optimal revenue against any distribution within this class. Furthermore, a second price auction is near-optimal when the class of admissible distributions is that of those with increasing virtual values (aka regular). Under this class, it guarantees a fraction of 50% of optimal revenues and no mechanism can guarantee more than 55.6%. (joint work with Amine Allouah)

Link to paper: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3124738

Bio:

Omar Besbes is an Associate Professor in the Decision, Risk, & Operations division at the Graduate School of Business, Columbia University. His primary research interests are in the area of data-driven decision-making, with a focus on applications in pricing and revenue management, e-commerce, online advertising, and service systems. His research has been recognized by various prizes including the 2012 INFORMS Revenue Management and Pricing Section prize, the 2013 M&SOM best paper award and the 2017 M&SOM young scholar award. He serves on the editorial boards of Operations Research and Management Science.

Omar is a graduate of Ecole Polytechnique (France) and received a M.Sc. from Stanford University in 2000 and a Ph.D. from Columbia University in 2008. Before joining Columbia, he was on the faculty at the Wharton School, University of Pennsylvania.

Subvector Inference in Partially Identified Models with Many Moment Inequalities

Abstract:

In this work we consider bootstrap-based inference methods for functions of the parameter vector in the presence of many moment inequalities where the number of moment inequalities, denoted by p, is possibly much larger than the sample size n. In particular this covers the case of subvector inference, such as the inference on a single component associated with a treatment/policy variable of interest. We consider a min-max of (centered and non-centered) Studentized statistics and study the properties of the associated critical values. In order to establish that we provide a new finite sample analysis that does not rely on Donsker's properties and establish new central limit theorems for the min-max of the components of random matrices. Furthermore, we consider the anti-concentration properties of the min-max of the components of a Gaussian matrix and propose bootstrap based methods to estimate them. In turn this provides a valid data-driven to set the tuning parameters of the bootstrap-based inference methods. Importantly, the tuning parameters generalize choices of literature for Donsker's classes (and showing why those would not be appropriate in our setting) which might better characterize finite sample behavior. This is co-authored with Federico Bugni and Victor Chernozhukov.

Bio:

Alexandre Belloni is a Professor at Duke University. He received his Ph.D. in Operations Research from the Massachusetts Institute of Technology (2006) and a M.Sc. in Mathematical Economics from IMPA (2002). He deferred the offer to join the faculty at Duke University to accept the IBM Herman Goldstein Postdoctoral Fellowship (2006-2007). Professor Belloni’s research interests are on econometrics, statistics and optimization. He received the 2007 Young Researchers Competition in Continuous Optimization Award. His research papers have appeared in journals such as Econometrica, Review of Economic Studies, Annals of Statistics, Marketing Science, Management Science and Operations Research. He serves as associate editor for different journals and is currently the Area Editor for Machine Learning and Data Science at Operations Research.

Multidimensional Text Mining with Limited Supervision

Abstract:

Unstructured text, as one of the most important data forms, plays a crucial role in domains such as cybersecurity, healthcare informatics, and cyber-physical systems. In many emerging applications, people's information need from text data is becoming multidimensional---they demand useful insights along multiple aspects from the given text corpus. However, acquiring multidimensional knowledge from massive text data challenges existing data mining techniques. In this talk, I will present a structuring-and-mining framework for facilitating acquiring multidimensional knowledge from text data. It organizes unstructured text into a multidimensional and multi-granular structure, from which end users can easily select relevant data with declarative queries and apply any data mining primitives thereafter. I will detail two core algorithms in this framework, including (1) a weakly supervised text classification algorithm; and (2) an abnormal event detection algorithm. The algorithms in the framework all require little supervision and are thus particularly appealing in scenarios where labeled data are expensive to acquire.

Bio:

Chao Zhang is an Assistant Professor at College of Computing, Georgia Institute of Technology. His research area is data mining and machine learning. He is particularly interested in developing label-efficient and robust learning techniques, with applications in text mining and spatiotemporal data mining. Chao has published more than 40 papers in top-tier conferences and journals, such as KDD, WWW, SIGIR, VLDB, and TKDE.  He is the recipient of the ECML/PKDD Best Student Paper Runner-up Award (2015) and the Chiang Chen Overseas Graduate Fellowship (2013). Before joining Georgia Tech, he obtained his Ph.D. degree in Computer Science from University of Illinois at Urbana-Champaign in 2018.

We strongly encourage students to act now to seek full-time employment, internships, and projects (rather than waiting until the end of the semester). Plus, enjoy a free pizza lunch!

EVENT DETAILS

Where: ISyE Main Bldg, 2nd Floor Atrium

When: Tuesday, March 12 | 11am - 2pm

What: The session will include:

• Networking opportunities in the ISyE atrium
• Food and refreshments

Please plan on staying for the duration of the event and bring copies of your resume and business cards. Dress is business casual.

REGISTER ONLINE by noon March 7th and upload your resume to be forwarded to the organization representatives!

EVENT SPONSOR

The event is sponsored through the generosity and support of JP Morgan Chase & Co. and APICS - Atlanta Chapter. APICS is a non-profit educational organization addressing operations management and supply chain management issues, and providing professional development opportunities to its members. APICS Membership is free for full time students. Join today at www.apics.org/join and start networking at local APIC Atlanta events. Also make sure to stop by the APICS table at the event.

This course will demonstrate how warehouse operations are a key enabler to a successful supply chain implementation and the starting point for a transformation. It is critical to manage safety, quality and efficiency. Learn to leverage the lean supply chain modifications to improve customer responsiveness and reduce operating costs and in doing so contributing to a supply chain that creates a competitive advantage for a company. To accomplish this goal, we must bring lean principles into the warehouse and distribution center.

WHO SHOULD ATTEND

Supply chain professionals, logistics professionals, material managers, production control managers, transportation managers, warehousing managers and purchasing managers

HOW YOU WILL BENEFIT

Upon completion of this course, you will be able to:

• Isolate the key elements of lean thinking to be used in the warehouse
• Value stream map warehouse operations
• Utilize lean tools to reduce waste in the warehouse
• Create a warehouse operation based on visual management and real time problem solving
• Reduce inventories in warehouse operations
• Create collaboration between warehousing and other functional areas

Benefits:

• Reduce inventories in the warehouse
• Reduce warehousing costs by as much as 25%
• Implement lean in the warehouse
• Create logistics collaboration between warehousing and other functional areas

WHAT IS COVERED

• Lean Warehouse Overview
• Supply Chain Implementation Framework
• Lean Storage Planning Approach
• Application of a Lean Storage Location Sizing Method
• JIT Implementation Approach
• How To Develop Standard Work Batches
• Generation of an Operational Diagram
• Creation of a Daily Operational Work Load Plan
• Development of a Progress Control Board

COURSE MATERIALS

Required

• Laptop
• Calculator

Provided

• Memory stick
• Book: Managing Global Supply & Risk: Best Practices Concepts and Strategies by Dr. Robert Trent and Lew Roberts

COURSE PREREQUISITES

None.

CERTIFICATE INFORMATION

This course is part of the Distribution Operations Analysis and Design (DOAD) Certificate.

PROGRAM TIMES

On the first day, please check in at least 30 minutes before the class start time.

• First Day (8:00am-5:00pm)
• Second Day (8:00am-5:00pm)
• Third/Last Day (8:00am-5:00pm)

Emerging urban logistics applications need to address a wide range of challenges, including complex traffic conditions and time-sensitive requirements. In this paper, in the context of urban logistics we consider a vehicle routing problem with time-dependent travel times and time windows (VRPTW), and the goal is to minimize the total generalized costs including the transportation cost, vehicular waiting cost and the fixed cost associated with each vehicle,  as well as capturing the endogenous traffic patterns. We adopt a high-dimensional space-time network flow model to formulate the underlying vehicle routing problem with a rich set of criteria and constraints. A thorny issue, when solving VRP with side constraints, is how to iteratively improve both primal and dual solution quality in general and how to break symmetry due to many identical solutions, especially with homogeneous vehicles. Along this line, many coupling constraints, such as the consensus constraints across different agents or decision makers, need to be carefully addressed to find high-quality optimal or close-to-optimal solutions under medium-scale or large-scale instances. Currently, Alternating Direction Method of Multipliers (ADMM) is widely used in the field of convex optimization, as an integration of Augmented Lagrangian relaxation and block coordinate descent methods for a large number of machine learning and large-scale continuous system optimization and control. In this paper, we first introduce the ADMM to the multi-vehicle routing problem, which is a special case of integer linear programming, and demonstrate the way to reduce the quadratic penalty terms used in ADMM into simple linear functions. In a broader context, a computationally reliable dual decomposition framework is developed to iteratively improve both primal and dual solution quality. We examine the performance of the proposed approach using classical a number of VRP benchmark instances. A real-world instance is specifically tested, based on a problem solving competition offered by Jingdong Logistics, a major E-commerce company.

Bio:

Xuesong Zhou is an Associate Professor of Transportation Systems in the School of Sustainable Engineering and the Built Environment at Arizona State University (ASU) in Tempe, Arizona. Dr. Zhou’s research work focuses on dynamic traffic assignment, traffic estimation and prediction, large-scale routing and rail scheduling. Dr. Zhou is currently an Associate Editor of Transportation Research Part C, an Associate Executive Editor-in-Chief of Urban Rail Transit, an Associate Editor of Networks and Spatial Economics, an Editorial Board Member of Transportation Research Part B. He was the formal Chair of INFORMS Rail Application Section (2016), and the Co-Chair of the IEEE ITS Society Technical Commit­tee on Traffic and Travel Management, as well as a subcommittee chair of the TRB Committee on Transportation Network Modeling (ADB30). He has published more than 50 papers related to dynamic traffic assignment and rail operations research in many journals with an H-index of 33.

Abstract: A general model of cyber risk at a given time in a specified organization is presented as the Bayesian analysis of a one-move game. Three illustrations reflect some aspects of the formulation of the risk model. The first one starts with the statistical analysis of a large database in a space organization and extends that analysis to include the risk of attacks that have not happened yet by a Bayesian analysis of extreme value scenarios. The second illustration focuses on the optimization of the level of connectivity between the cloud and a system such as a smart power grid, then extends the problem to include the optimal allocation of personnel between “watchers” of known vulnerabilities and “hunters” of new ones. The third model focuses on warnings of cyberattacks given the process by which “robots” that monitor the traffic may decide to cut connections when the cyber risk reaches a certain level, then may decide automatically to pass the baton to a human being when the threat exceeds the automatic monitoring system’s threshold of capabilities. The data used in that third illustration were collected in part through a number of “honeypots” placed around the world and the large number of attacks that they collected.

Bio: M. Elisabeth Paté-Cornell is a professor and the founding chair of the management science and engineering department at Stanford. Her specialty is engineering risk analysis, with applications to complex systems (space, medical, offshore platforms, cyber security, nuclear, etc.). She is a member of the National Academy of Engineering, the French Académie des Technologies, the Naval Post-Graduate School Advisory Board, and the NASA Advisory Council. She was a member of the President’s Foreign Intelligence Advisory Board (2001-2008) and several other boards. She holds a B.S. in mathematics and physics from Marseille (France), an engineering degree in applied math/CS from the Institut Polytechnique de Grenoble (France), and an M.S. in operations research and a Ph.D. in engineering-economic systems, both from Stanford University.

Reception immediately following in the Klaus Atrium

Abstract:

The objective of this research is to develop machine learning methods that predictively improve the outcome of design solutions through the acquisition, fusion and mining of large-scale, publicly-available data. It has been reported that 70-80% of the manufacturing costs of a product are determined during the design phase. Towards enhancing the efficiency of the design process and creating personalized design solutions, our research focuses on three core thrusts:

Research Thrust 1: Feature Discovery and Quantification: Develop machine learning algorithms that mine large scale, publicly-available data in order to identify and quantify relevant design features.

Research Thrust 2: Scalable Generative Design: Develop machine learning models (currently based on Deep Neural Network principles) that result in the automatic, scalable generation of design concepts (both 2D sketches and 3D CAD models).

Research Thrust 3: Scalable Design Evaluation: Develop design evaluation methods (currently based on Deep Reinforcement Learning principles) that result in the automatic evaluation of design concepts, based on computer-learned physics of an environment.

The fundamental concepts of machine learning-driven design extend well beyond consumer products and include the design of more efficient manufacturing processes (e.g., machine learning models for improving the efficiency of additive manufacturing processes), healthcare systems (e.g., machine learning models for early detection, as well as long-term management of patients’ health-related abnormalities), and educational experiences (e.g., advancing personalized learning through adaptive machine learning models).

Bio:

Dr. Conrad Tucker holds a joint appointment as Associate Professor in Engineering Design and Industrial and Manufacturing Engineering at The Pennsylvania State University. He is also affiliate faculty in Computer Science and Engineering. Dr. Tucker is the director of the Design Analysis Technology Advancement (D.A.T.A) Laboratory. His research focuses on the design and optimization of systems through the acquisition, integration and mining of large scale, disparate data.

Dr. Tucker has served as PI/Co-PI on federally/non-federally funded grants from the National Science Foundation (NSF), the Air Force Office of Scientific Research (AFOSR), the Defense Advanced Research Projects Agency (DARPA), the Army Research Laboratory (ARL), the Office of Naval Research (ONR) through the NSF Center for e-Design, and most recently, the Bill and Melinda Gates Foundation (BMGF). He is currently serving as PI and Site Director of the NSF Center for Health Organization Transformation (CHOT), an NSF Industry/University Cooperative Research Center at Penn State. In February 2016, he was invited by National Academy of Engineering (NAE) President Dr. Dan Mote, to serve as a member of the Advisory Committee for the NAE Frontiers of Engineering Education (FOEE) Symposium. Dr. Tucker is the recipient of the American Society of the Engineering Education’s (ASEE) Summer Faculty Fellowship Program (SFFP) award and conducted research at the Air Force Institute of Technology at the Wright Patterson Air Force Base during Summer 2014 and Summer 2015. He received his Ph.D., M.S. (Industrial Engineering), and MBA degrees from the University of Illinois at Urbana-Champaign, and his B.S. in Mechanical Engineering from Rose-Hulman Institute of Technology. Dr. Tucker is part of the inaugural class of the Gates Millennium Scholars (GMS) program, funded by a $1 billion grant from the Bill and Melinda Gates Foundation.

In this work we consider bootstrap-based inference methods for functions of the parameter vector in the presence of many moment inequalities where the number of moment inequalities, denoted by p, is possibly much larger than the sample size n. In particular this covers the case of subvector inference, such as the inference on a single component associated with a treatment/policy variable of interest. We consider a min-max of (centered and non-centered) Studentized statistics and study the properties of the associated critical values. In order to establish that we provide a new finite sample analysis that does not rely on Donsker's properties and establish new central limit theorems for the min-max of the components of random matrices. Furthermore, we consider the anti-concentration properties of the min-max of the components of a Gaussian matrix and propose bootstrap based methods to estimate them. In turn this provides a valid data-driven to set the tuning parameters of the bootstrap-based inference methods. Importantly, the tuning parameters generalize choices of literature for Donsker's classes (and showing why those would not be appropriate in our setting) which might better characterize finite sample behavior. This is co-authored with Federico Bugni and Victor Chernozhukov.

Bio:

Alexandre Belloni is a Professor at Duke University. He received his Ph.D. in Operations Research from the Massachusetts Institute of Technology (2006) and a M.Sc. in Mathematical Economics from IMPA (2002). He deferred the offer to join the faculty at Duke University to accept the IBM Herman Goldstein Postdoctoral Fellowship (2006-2007). Professor Belloni’s research interests are on econometrics, statistics and optimization. He received the 2007 Young Researchers Competition in Continuous Optimization Award. His research papers have appeared in journals such as Econometrica, Review of Economic Studies, Annals of Statistics, Marketing Science, Management Science and Operations Research. He serves as associate editor for different journals and is currently the Area Editor for Machine Learning and Data Science at Operations Research.

Statistical Modeling and Optimization for Optimal Adaptive Trial Design in Personalized Medicine

Abstract:

We provide a new modeling framework and adopt modern optimization tools to attack an important open problem in statistics. In particular, we consider the optimal adaptive trial design problem in personalized medicine. Adaptive enrichment designs involve preplanned rules for modifying enrollment criteria based on accruing data in a randomized trial. We focus on designs where the overall population is partitioned into two predefined subpopulations, e.g., based on a biomarker or risk score measured at baseline for personalized medicine. The goal is to learn which populations benefit from an experimental treatment. Two critical components of adaptive enrichment designs are the decision rule for modifying enrollment, and the multiple testing procedure. We provide a general framework for simultaneously optimizing these components for two-stage, adaptive enrichment designs through Bayesian optimization. We minimize the expected sample size under constraints on power and the familywise Type I error rate. It is computationally infeasible to directly solve this optimization problem due to its nonconvexity and infinite dimensionality. The key to our approach is a novel, discrete representation of this optimization problem as a sparse linear program, which is large-scale but computationally feasible to solve using modern optimization techniques. Applications of our approach produce new, approximately optimal designs. In addition, we shall further discuss several extensions to solve other related statistical problems.

Bio: 

Ethan is an assistant professor at Penn State University. Before joining Penn State, he got his PhD from Princeton University in 2016 and his bachelor's degree from the National University of Singapore in 2010. He works on different problems such as statistical learning, high-dimensional inference and adaptive trial design from both statistical and computational perspectives. He won numerous awards in statistics and optimization such as Best Paper Prize for Young Researchers in Continuous Optimization (jointly with Mengdi Wang and Han Liu), ENAR Distinguished Student Paper Prize and IMS Travel Award. 

Abstract: Although “big data” is ubiquitous in data science, one often faces challenges of “small data,” as the amount of data that can be taken or transmitted is limited by technical or economic constraints. To retrieve useful information from the insufficient amount of data, additional assumptions on the signal of interest are required, e.g. sparsity (having only a few non-zero elements). Conventional methods favor incoherent systems, in which any two measurements are as little correlated as possible. In reality, however, many problems are coherent.  I will present two nonconvex approaches: one is the difference of the L1 and L2 norms and the other is the ratio of the two. The difference model works particularly well in the coherent regime, while the ratio is a scale-invariant metric that works better when underlying signals have large fluctuations in non-zero values. Various numerical experiments have demonstrated advantages of the proposed methods over the state-of-the-art. Applications, ranging from super-resolution to low-rank approximation, will be discussed.

Bio: Yifei Lou has been an Assistant Professor in the Mathematical Sciences Department, University of Texas Dallas, since 2014. She received her Ph.D. in Applied Math from the University of California Los Angeles (UCLA) in 2010. After graduation, she was a postdoctoral fellow at the School of Electrical and Computer Engineering Georgia Institute of Technology, followed by another postdoc training at the Department of Mathematics, University of California Irvine from 2012-2014. Her research interests include compressive sensing and its applications, image analysis (medical imaging, hyperspectral, imaging through turbulence), and (nonconvex) optimization algorithms.

Adapting Maximum Likelihood Theory in Modern Applications

Abstract:

Maximum likelihood estimation (MLE) is influential because it can be easily applied to generate optimal, statistically efficient procedures for broad classes of estimation problems. Nonetheless, the theory does not apply to modern settings—such as problems with computational, communication, or privacy considerations—where our estimators have resource constraints. In this talk, I will introduce a modern maximum likelihood theory that addresses these issues, focusing specifically on procedures that must be computationally efficient or privacy-preserving. To do so, I first derive analogues of Fisher information for these applications, which allows a precise characterization of tradeoffs between statistical efficiency, privacy, and computation. To complete the development, I also describe a recipe that generates optimal statistical procedures (analogues of the MLE) in the new settings, showing how to achieve the new Fisher information lower bounds.

Bio:

Feng Ruan is a fifth year Ph.D. student in the Department of Statistics at Stanford University, advised by Prof. John Duchi. He is broadly interested in developing theory and algorithm for inference under resource constraints, for stochastic convex and nonconvex optimization, and for high dimensional statistics. He is a recipient of the E.K. Potter Stanford Graduate Fellowship from Stanford University.

Big data, Google, and infectious disease prediction: A statistical perspective

Abstract:

Big data generated from the Internet present a great opportunity for real-time disease surveillance and tracking, such as flu in the United States or dengue fever in tropical countries. These big-data insights into real-time infectious disease prediction could help public health officials make timely decisions to save lives. We proposed a theoretically rooted method that leads to robust and accurate real-time tracking of infectious diseases. Our method significantly outperforms all previous internet-based tracking models, including Google Flu Trends and Google Dengue Trends.

In the case of flu, we introduced our real-time digital flu detection method ARGO (AutoRegressive with GOogle data), which combines time series information with Google search data. ARGO is derived from a hidden Markov structure of data-generating mechanism. With a sliding two-year window and an L1 penalty for training, ARGO can incorporate new information as it becomes available, and can automatically select or adjust the most useful Google search queries. We extended ARGO to track dengue fever with great success in five tropical countries including Brazil, Mexico, Thailand, Singapore, and Taiwan. ARGO is then further extended to incorporate cloud-based electronic health records and to generate near-future predictions weeks ahead. Our latest development upgrades the method for infectious disease tracking in spatial scale. Thanks to the ubiquity of internet search data, ARGO is now capable of real-time disease tracking not only at national level but also at regional level. The upgraded ARGO uses penalized spatial-temporal information pooling, making it flexible, self-correcting, robust and scalable.

Bio:

Shihao Yang is a PhD candidate at Department of Statistics from Harvard University, advised by Prof. Samuel Kou. His primary research interest is to harness the power of big data to solve real-life problems, with focus on three perspectives: methodological development, computational tools, and probabilistic modeling.

On the methodological development perspective, he developed methods for infectious disease forecast based on internet search data, and built a tailor-made matching method to study cancer immunotherapy with electronic health data. On the computation perspective, he introduced a new method for parallelizable Markov chain Monte Carlo, and another fast approximation method for inference in dynamic systems via constrained Gaussian process. On the probabilistic modeling perspective, he proposed novel stochastic differential equations to capture the underlying dynamics of high-speed, high-volume financial market transactions.

Bowen Hua and Ross Baldick, University of Texas at Austin

Unit commitment and dispatch of generation in electricity markets involves the ISO sending target quantity instructions to each generator.  Ideally, energy prices provide incentives for profit maximizing market participants to comply with efficient commitment and dispatch instructions in the short-term, and to develop new generation (based on anticipation of future energy prices) at the right place and time over the long-term. Various issues prevent this ideal, including the non-convexity of the underlying unit commitment problem, so that start-up and no-load costs of generation units may not be covered by sales of energy at locational marginal prices.  To encourage generators to comply with commitment and dispatch, non-negative profit is guaranteed by the ISO, but this guarantee necessitates non-anonymous “uplift” payments. Because energy prices alone do not provide all the information about profitability in the market, incentives for new generation to enter the market are weakened.  Moreover, the energy prices do not reveal to demand the full cost of providing energy, weakening incentives for demand response.  Both issues arguably contribute to resource adequacy problems. Convex hull pricing is a uniform pricing scheme that minimizes the amount of uplift payments. The Lagrangian dual problem of the unit commitment and economic dispatch problem has been used to determine the convex hull prices. This approach is computationally expensive for market implementations. We propose a convex programming formulation for convex hull pricing that is polynomially solvable. We describe explicitly the convex hull of individual units’ feasible commitment and dispatch decisions. We show that our formulation leads to the exact convex hull prices when ramping constraints are not present, and that the exactness is preserved when ancillary services or any linear system-wide constraints (such as the transmission constraints) are considered. When ramping constraints are considered, an exponential number of valid inequalities are needed to describe the convex hulls. In this case, we use a tractable approximation. Numerical tests are conducted on several examples found in the literature.  Although this work provides for convenient implementation of convex hull pricing, significant issues remain.  For example, the incentives for generators to reveal truthful start-up and no-load information are not well understood under either the traditional uplift approach or convex hull pricing.  Moreover, under convex hull prices, profit maximizing generators have incentives to deviate from the ISO-determined dispatch instructions.

Dr. Banzhaf will provide an overview of the Atlanta Federal Statistical Research Data Center (FSRDC), one of 29 FSRDCs across the country.  The Atlanta FSRDC provides access to numerous internal government-sponsored datasets, including restricted-use versions of the American Community Survey and Decennial Census, the Economic Census, UMetrics data, the Census of Fatal Occupational Injuries, and many other economic and demographic datasets.  With a special focus on projects of interest to statistical researchers, Dr. Banzhaf will explain the advantages of using restricted data, the process for developing a successful proposal to use the data, and the logistics of conducting research in the Atlanta FSRDC.  For more information, please contact melissa.r.banzhaf@census.gov or visit the FSRDC website: https://www.census.gov/fsrdc.

Bio:

Dr. Banzhaf received a BA in economics from Duke University and a PhD in economics from the University of North Carolina at Chapel Hill.  She has served as a Census Bureau economist and the administrator of the Atlanta FSRDC since it opened in 2011.  Her current research focuses on the economics of the family and the long-term economic outcomes of air pollution exposure.  Prior to working for the U.S. Census Bureau, she held positions with the U.S. Bureau of Labor Statistics in Washington, DC, and Triangle Economic Research and RTI International in Durham, NC. 

Abstract: 

Electricity goals around the world tend to focus on increasing social benefit through one of two things: (1) increasing overall system sustainability or (2) increasing access to electricity. In pursuit of these goals decision makers will need modeling tools that can inform decisions, in a way that is flexible enough to include a wide range of preferences and goals. It is clear that the future generation mix of the power system will change, but the most sustainable solution will change based on a country's goals.

This talk will explore various options for power grid expansion in both developed and developing countries. I present two papers that focus on approaches to power system planning that help decision makers reach their energy targets. The first part is focused on evaluating the sustainability of a set of generation portfolios. We take a multi-model approach, first determining the reliability of the system overall, then evaluating different generation portfolios based on seven sustainability criteria. The second part focuses on the UN Sustainable Development Goals, and energy access in developing countries. We have developed a utility maximization framework that incorporates stakeholder equity preferences into an electricity planning model. This can be used by decision makers to determine the best method of grid expansion to meet electricity access goals subject to system and budget constraints.

Bio:

Destenie Nock is a Ph.D. Candidate in Industrial Engineering and Operations Research at the University of Massachusetts Amherst, an NSF Graduate Research Fellow, and an Offshore Wind Energy IGERT Fellow. Prior to this she was a Mitchell Fellow in Northern Ireland where she earned an MSc in Leadership for Sustainable Development at Queens University of Belfast. Her research is focused on applying optimization and decision analysis tools to evaluate the sustainability and reliability of the electricity grid in disparate energy systems. She has two B.S. degrees from North Carolina A&T State University in Electrical Engineering and Applied Mathematics.  In addition, she has held internships at Argonne National Lab, the Utility Regulator of Northern Ireland, and Exxon Mobil

Learning to Make Decisions in Statistical and Strategic Environments

Abstract:

Data-driven decision making, lying at the intersection between learning and decision making, has emerged as an important paradigm in operations research and data science. In the first part of the talk, we focus on data-driven decision making in a statistical environment. In many settings, a decision-maker wishes to learn a rule, or policy, that maps from observable characteristics of an individual to an action. Examples include selecting offers, prices, advertisements, or emails to send to consumers, as well as the problem of determining which medication to prescribe to a patient. Here we study the offline multi-action policy learning problem with observational data and where the policy may need to respect budget constraints or belong to a restricted policy class such as decision trees. We build on the theory of efficient semi-parametric inference in order to propose and implement a policy learning algorithm that achieves asymptotically minimax-optimal regret. To the best of our knowledge, this is the first result of this type in the multi-action setup, and it provides a substantial performance improvement over the existing learning algorithms. We then discuss optimization schemes for implementing decision-tree based policies.

In the second part of the talk, we consider a model of multi-agent online strategic decision making, in which the reward structures of agents are given by a general continuous game. After introducing a general equilibrium stability notion for continuous games, called variational stability, we examine the well-known online mirror descent (OMD) learning algorithm and show that OMD converges to variationally stable Nash equilibria. Subsequently, by developing various algorithmic variants, we show that convergence to Nash equilibria still holds even in the presence of severely imperfect information, including noise, (fully asynchronous and unbounded) delays and loss. We also discuss two applications of this theoretical framework: one in distributed stochastic optimization and the other in power management.

Bio: 

Zhengyuan Zhou is a PhD candidate in the Information System Laboratory in the Department of Electrical Engineering at Stanford University, with a master in statistics, a master in computer science and a master in economics, all from Stanford University. He obtained his B.E. in Electrical Engineering and Computer Sciences and B.S. in Mathematics from UC Berkeley. He has had industry experience in Google, Microsoft Research and Oracle. His research interests include machine learning, stochastic optimization, applied probability and game theory.

Preference elicitation for participatory budgeting

Abstract:

Participatory budgeting enables the allocation of public funds by collecting and aggregating individual preferences; it has already had a sizable real-world impact. But making the most of this new paradigm requires a rethinking of some of the basics of computational social choice, including the very way in which individuals express their preferences. We analytically compare four preference elicitation methods — knapsack votes, rankings by value or value for money, and threshold approval votes — through the lens of implicit utilitarian voting, and find that threshold approval votes are qualitatively superior. This conclusion is supported by experiments using data from real participatory budgeting elections, while a user study  examines the practical implications of the respective methods.

Bio:

Gerdus is a PhD student in operations research at Carnegie Mellon University; he is advised by Ariel Procaccia and John Hooker. His research interests include computational social choice, fair division and discrete optimization. His work emphasizes the impact of human factors like bias, incomplete information and various notions of fairness on traditional voting and assignment problems. 

We strongly encourage students to act now to seek full-time employment, internships, and projects (rather than waiting until the end of the semester). Plus, enjoy a free pizza lunch!

EVENT DETAILS

Where: ISyE Main Bldg, 2nd Floor Atrium

When: Wednesday, January 23 | 11am - 2pm

What: The session will include:

• Networking opportunities in the ISyE atrium
• Food and refreshments

Please plan on staying for the duration of the event and bring copies of your resume and business cards. Dress is business casual.

REGISTER ONLINE by midnight January 15th and upload your resume to be forwarded to the organization representatives!

EVENT SPONSOR

The event is sponsored through the generosity and support of JP Morgan Chase & Co. and APICS - Atlanta Chapter. APICS is a non-profit educational organization addressing operations management and supply chain management issues, and providing professional development opportunities to its members. APICS Membership is free for full time students. Join today at www.apics.org/join and start networking at local APIC Atlanta events. Also make sure to stop by the APICS table at the event.

We look forward to seeing you – and gaining insights on how GT SCL can assist in community research, education, and innovation needs. 

Please RSVP at https://welovelogistics.eventbrite.com and contact timbrown@gatech.edu with questions.

Bridging MCMC and Optimization

Abstract:

In this talk, I will discuss three ingredients of optimization theory in the context of MCMC: Non-convexity, Acceleration, and stochasticity. I will focus on a class of non-convex objective functions arising from mixture models. For that class of objective functions, I will demonstrate that the computational complexity of a simple MCMC algorithm scales linearly with the model dimension, while optimization problems are NP-hard.

I will then study MCMC algorithms as optimization over the KL-divergence in the space of measures. By incorporating a momentum variable, I will discuss an algorithm which performs accelerated gradient descent over the KL-divergence. Using optimization-like ideas, a suitable Lyapunov function is constructed to prove that an accelerated convergence rate is obtained.

Finally, I will present a complete recipe for constructing stochastic gradient MCMC algorithms that translates the task of finding a valid sampler into one of  choosing two matrices. I will then describe how stochastic gradient MCMC algorithms can be applied to applications involving temporally correlated data, where the challenge arises from the need to break the dependencies when considering minibatches of observations.

Bio: