All attendees receive free access to the INFORMS 2017 TutORials in Operations Research online content concurrently with the meeting. Registrants of the 2017 INFORMS Annual Meeting have online access to the 2017 chapters, written by select presenters, beginning on October 21, 2017. Access this content using the link provided to all attendees by email or, if you are a 2017 member, simply login to INFORMS PubsOnLine.
The TutORials in Operations Research series is published annually by INFORMS as an introduction to emerging and classical subfields of operations research and management science. These chapters are designed to be accessible for all constituents of the INFORMS community, including current students, practitioners, faculty, and researchers. The publication allows readers to keep pace with new developments in the field, and serves as augmenting material for a selection of the tutorial presentations offered at the INFORMS Annual Meetings.
8 – 9:30am
An Introduction to Two-Stage Stochastic Mixed-Integer Programming
Department of Industrial and Systems Engineering, University of Washington
Simge Küçükyavuz is an Associate Professor in the Industrial and Systems Engineering Department at the University of Washington. Prior to joining U. Washington, she was a faculty member at the Ohio State University and the University of Arizona and a research associate at Hewlett-Packard Laboratories. She received her Ph.D. degree in Industrial Engineering and Operations Research from the University of California, Berkeley. Her interests are in mixed-integer programming, large-scale optimization, optimization under uncertainty, and their applications. Her research is supported by multiple grants from the National Science Foundation, including the 2011 CAREER Award. She is the co-winner of the 2015 ICS (INFORMS Computing Society) Prize, and serves on the editorial boards of several journals.
Epstein Department of Industrial and Systems Engineering, University of Southern California
Suvrajeet Sen is a Professor at the Daniel J. Epstein Department of Industrial and Systems Engineering at the University of Southern California. Prior to joining USC, he was a Professor at Ohio State University, and University of Arizona. He has also served as the Program Director at NSF. Professor Sen’s research is devoted to many categories of optimization models, and he has published over a hundred papers, with the vast majority of them dealing with models, algorithms and applications of Stochastic Programming problems. He has served on several editorial boards, including Operations Research as Area Editor for Optimization. Professor Sen was instrumental in founding the INFORMS Optimization Society and also served as its Chair recently. Professor Sen is a Fellow of INFORMS, and also led a group that was awarded the INFORMS Computing Society Prize in 2015.
This paper provides an introduction to algorithms for two-stage stochastic mixed integer programs. Our focus is on methods which decompose the problem by scenarios representing randomness in the problem data. The design of these algorithms depend on where the uncertainty appears (right-hand-side, recourse matrix and/or technology matrix) and where the continuous and discrete decision variables are (first-stage and/or second-stage). In addition we provide computational evidence that, similar to other classes of stochastic programming problems, decomposition methods can provide desirable theoretical properties (such as finite convergence) as well as enhanced computational performance when compared to solving a deterministic equivalent formulation using an advanced commercial MIP solver.
11am – 12:30pm
Operations Research Approaches for Building Demand Response in a Smart Grid
GERAD & Polytechnique Montreal, Canada
Miguel F. Anjos is a Full Professor in the Department of Mathematics and Industrial Engineering of Polytechnique Montreal, where he holds the NSERC-Hydro-Quebec-Schneider Electric Senior Industrial Research Chair on Optimization for the Smart Grid. His research interests are in optimization theory and algorithms with applications to power systems management and smart grid design. He is the Founding Academic Director of the Trottier Institute for Energy at Polytechnique from 2013 to 2016, Editor-in-Chief of Optimization and Engineering and serves on several other editorial boards. He was elected to three-year terms on the Council of the Mathematical Optimization Society and as Program Director for the SIAM Activity Group on Optimization, and to a two-year term as Vice-Chair of the INFORMS Optimization Society. He serves on the Mitacs Research Council since its creation in 2011. His allocades include a Canada Research Chair, the Meritas Teaching Award, a Humboldt Research Fellowship, and the Queen Elizabeth II Diamond Jubilee Medal. He is a fellow of the Canadian Academy of Engineering.
GERAD & Polytechnique Montreal, Canada
Electric power systems need to ensure that production and demand of electricity are continuously in balance. With fundamental changes taking place in the power grids of many countries due to a variety of technological and policy developments, there is a need to obtain additional flexibility to achieve this essential power balance. Demand response refers to the collection of all the means to obtain this flexibility from the demand side of the balance. We present a selection of contributions of operations research to the provision of demand response by the residential, commercial and institutional sectors of the economy. The aspects covered include electricity tariffs, building energy management systems, load estimation, local generation, electric vehicles, energy storage, and building-level aggregation. We conclude with a brief discussion of current opportunities for operations research to support the development and realization of the potential of demand response.
1:30 – 3pm
Easy Affine Markov Decision Processes: Properties and Applications
This tutorial introduces a class of decomposable a ne Markov decision processes (MDPs) that have continuous multi-dimensional endogenous states and actions, and an exogenous state that follows an exogenous Markov chain. We show that, unlike most MDPs with continuous state and actions, decomposable affine MDPs are free of the curse of dimensionality and can be solved easily and exactly. These nice properties are attributed to its a_ne dynamics and a_ne single-period rewards, its decomposable action space, and the polyhedral features of the decomposed action space. Exploiting its structure, we demonstrate that a decomposable a_ne MDP with a finite-horizon criterion has a value function that is affine in the endogenous state, and has an extremal optimal policy; and the value function and the extremal optimal policy are determined by the solution of a set of auxiliary equations. At the end of the tutorial, we illustrate the potential applicability of decomposable affine MDPs using examples of fishery management and dynamic capacity portfolio management.
4:30 – 6pm
Metabolic Networks and Modern Research Problems in Operations Research
Virgina Commonwealth University, Departments of Statistical Sciences and Operations Research and Supply Chain Management and Analytics
J. Paul Brooks is an associate professor in the Department of Statistical Sciences and Operations Research and the Department of Supply Chain Management and Analytics at Virginia Commonwealth University. His research involves the use of optimization in developing new methods for data analysis and their application, and in other areas of applied optimization. Application areas include human microbiome data, computational biology, and scheduling. He is a member of the Council of the INFORMS Section on Data Mining and is active in the INFORMS Computing Society.
Rose-Hulman Institute of Technology, Department of Mathematics
Allen Holder is a professor of mathematics at the Rose-Hulman Institute of Technology. His research interests lie in mathematical programming and its applications to healthcare and biology. He was the recipient of the 2000 Pierskalla Award for his original work on the optimal design of radiotherapy treatments, and he received the 2015 Outstanding Scholar Award from the Board of Trustees at Rose-Hulman. He serves as an area editor for the INFORMS Journal on Computing, covering applications in biology, medicine, and healthcare. He enjoys undergraduate research and has published with many of his undergraduate students.
Biochemical reactions that sustain life are modeled as a system of differential equations, forming a metabolic network. These nonlinear systems curtail into linear, algebraic systems under asymptotic assumptions, which thus define the feasible flux states of a cell in equilibrium. An optimization problem is then imposed to further identify metabolic conditions in an optimal growth situation. The result is a linear programming (LP) technique called Flux Balance Analysis (FBA), and the primary goal of this tutorial is to illustrate several modern research questions about FBA that are compliant with operations research. We pose several open avenues for research that touch LP, mixed-integer programming (MIP), nonlinear programming, combinatorial optimization, optimization under uncertainty, and robust optimization.
8 – 9:30am
Introduction to Disaggregate Demand Models
Transport and Mobility Laboratory, School of Architecture, Civil and Environmental Engineering,
Ecole Polytechnique Fédérale de Lausanne, Switzerland
Email: email@example.com, firstname.lastname@example.org
Michel Bierlaire is a full professor at the Ecole Polytechnique Fédérale de Lausanne (EPFL), where he serves as the director of TraCE, the Transportation Center, and the director of Doctoral Program in Civil and Environmental Engineering. He received a PhD in Mathematics from University of Namur, Belgium. After that, he worked as a research associate and project manager at the Intelligent Transportation Systems Program of MIT before he joined EPFL in 1998. His main expertise is in the design, development and applications of models and algorithms for the design, analysis and management of transportation systems. He is the founder of hEART: the European Association for Research in Transportation, the Editor-in-Chief of the EURO Journal on Transportation and Logistics, and served in several other advisory/editorial boards.
Demand information is an input for a great deal of operations research models. Assumed as given in many problem instances addressed in the literature, demand data are difficult to generate. In this tutorial, we provide an introduction to disaggregate demand models that are designed to capture in detail the underlying behavioral mechanisms at the foundation of the demand.
11am – 12:30pm
Department of Industrial Engineering, Bilkent University, 06800 Ankara, Turkey
Bahar Y. Kara is a Professor at the Department of Industrial Engineering at Bilkent University. Professor Kara works on Operations Research applications in logistics. She has published many papers dealing with models and applications of network optimization problems. She is one of founding coordinators, and an executive board member for the European Working Group on Humanitarian Operations (Euro-HOpe). She is currently active in humanitarian logistics, mainly on relief logistics and refugee movement.
Department of Industrial Engineering, Bilkent University, 06800 Ankara, Turkey
Many incidents requiring humanitarian assistance, which can be categorized mainly as disasters and long term crises, have taken place in the recent years. Humanitarian logistics play a crucial role in managing these events because of the response time constraints, limited resource amounts and high levels of uncertainty. Therefore, it is apparent that the challenges faced in humanitarian logistics differ from the conventional challenges encountered in commercial logistics applications such as having conflicting goals for different parties, funding and political issues. In this tutorial, we provide an introduction to humanitarian logistics by defining its characteristics, application areas and challenges. Then, we analyze and synthesize relief logistics and development logistics literature based on Operations Research (OR) problems encountered in disaster management cycle while putting emphasis on the last 10 years. For various disaster management stages; namely preparedness and response phases, certain case studies will be discussed using a real life dataset from Kartal region of Istanbul. Moreover, for development logistics, a case study on providing healthcare in rural areas will be evaluated.
1:30 – 3pm
Optimization Methods for Supervised Machine Learning: From Linear Models to Deep Learning
Lehigh University, Department of ISE
Lehigh University, Department of ISE
Katya Scheinberg is the Harvey E. Wagner Endowed Chair Professor at the Industrial and Systems Engineering Department at Lehigh University. After received her PhD degree in operations research from Columbia University, she had worked at the IBM T.J. Watson Research Center as a research staff member for over a decade before joining Lehigh in 2010. Katya’s main research areas are related to design and analysis of practical algorithms for continuous optimization problems such as derivative free optimization with applications in machine learning. In recent years, she has been focusing on large-scale optimization method for Big Data applications and Machine Learning. In 2015, jointly with Andy Conn and Luis Vicente, she received the Lagrange Prize awarded jointly by SIAM and MOS.
The goal of this tutorial is to introduce key models, algorithms, and open questions related to the use of optimization methods for solving problems arising in machine learning. It is written with an INFORMS audience in mind, specifically those readers who are familiar with the basics of optimization algorithms, but less familiar with machine learning. We begin by deriving a formulation of a supervised learning problem and show how it leads to various optimization problems, depending on the context and underlying assumptions. We then discuss some of the distinctive features of these optimization problems, focusing on the examples of logistic regression and the training of deep neural networks. The latter half of the tutorial focuses on optimization algorithms, first for convex logistic regression, for which we discuss the use of first-order methods, the stochastic gradient method, variance reducing stochastic methods, and second-order methods. Finally, we discuss how these approaches can be employed to the training of deep neural networks, emphasizing the difficulties that arise from the complex, nonconvex structure of these models.
4:40 – 6pm
Markov Decision Processes, AlphaGo, and Monte Carlo Tree Search: Back to the Future
Robert H. Smith School of Business & Institute for Systems Research, University of Maryland,
College Park, Maryland 20741, USA
Michael C. Fu holds the Smith Chair of Management Science in the Decision, Operations and Information Technologies department of the Robert H. Smith School of Business, with a joint appointment in the Institute for Systems Research and an affiliate appointment in the Department of Electrical & Computer Engineering (both in the Clark School of Engineering), all at the University of Maryland, College Park. He received degrees in math and EECS from MIT, Cambridge, MA, in 1985 and a Ph.D. in applied math from Harvard University in 1989. He is the co-author of the books, Conditional Monte Carlo: Gradient Estimation and Optimization Applications, which received the INFORMS Simulation Societys 1998 Outstanding Publication Award, and Simulation-Based Algorithms for Markov Decision Processes, and has also edited/co-edited four volumes: Perspectives in Operations Research, Advances in Mathematical Finance, Encyclopedia of Operations Research and Management Science (3rd edition), and Handbook of Simulation Optimization. He served as Department Editor for Stochastic Models and Simulation at Management Science, as Simulation Area Editor for OR, and on the Editorial Board of MOR, INFORMS Journal on Computing, IIE Transactions, and Production and Operations Management. He served as the OR Program Director at NSF from 2010–2012 and in 2015, and is a Fellow of IEEE and INFORMS.
In 2016, a computer Go-playing program called AlphaGo stunned the (human) world by winning a match (4 games to 1) against the reigning human world champion, a feat more impressive than previous victories by computer programs in chess (Deep Blue) and the TV game show Jeopardy (Watson). The main engine behind AlphaGo combines machine learning approaches in the form of deep neural networks with a technique called Monte Carlo tree search, whose roots can be traced back to an adaptive multistage sampling simulation-based algorithm for Markov decision processes (MDPs) published in Operations Research back in 2005 (and introduced even earlier in 2002). This tutorial describes AlphaGo and the simulation-based MDP algorithm, as well as providing contextual and historical background material for both, and uses simple examples to illustrate the main ideas behind Monte Carlo tree search.
7:30 – 9am
DII, Univeridad de Chile
Fernando Ordonez is a Professor in the Industrial Engineering Department at the University of Chile. He received his Mathematical Engineering degree, from the University of Chile in 1997 and his Ph.D. in Operations Research from MIT in 2002. His research focuses on mathematical optimization models, uncertainty, algorithms, and applications of optimization to engineering and management science. His research was awarded the Wagner prize for Excellence in Operations Research practice in 2012.
ISE, University of Southern California
Maged M. Dessouky is a Professor in Industrial and Systems Engineering at the University of Southern California and the Director of the Epstein Institute. He received B.S. and M.S. degrees from Purdue University and a Ph.D. in Industrial Engineering from the University of California, Berkeley. He is area/associate editor of Transportation Research Part B: Methodological, IIE Transactions, and Computers and Industrial Engineering, on the editorial board of Transportation Research Part E: Logistics and Transportation Review, and previously served as area editor of the ACM Transactions of Modeling and Computer Simulation and associate editor of IEEE Transactions on Intelligent Transportation Systems. He is a fellow of IIE and was awarded the 2007 Transportation Science and Logistics Best Paper Prize.
Recent technological developments such as GPS, mobile devices and increases in data storage and computation capacity have greatly enhanced the communication capabilities of travelers, facilitating ridesharing in real-time. This opportunity has created a new ridesharing industry built on making use of the unused vehicle capacity moving about on the roads. There are however still important barriers for widespread dynamic ridesharing adoption. It therefore becomes important to better understand the problems and challenges of ridesharing, creating the need for novel research that explicitly takes into account dynamic ridesharing. In this tutorial we review recent research on new vehicle routing models, cost sharing mechanisms, and planning models that incorporate ridesharing.
10:30am – 12pm
Quantitative Imaging System for Cancer Diagnosis and Treatment Planning: An Interdisciplinary Approach
School of Computing, Informatics, Decision Systems Engineering, Ira Fulton Schools of Engineering, Arizona State University
Email: Teresa.Wu@asu.edu, email@example.com, Yanzhe.Xu@asu.edu, firstname.lastname@example.org, email@example.com
Teresa Wu is a Professor of Industrial Engineering Program, School of Computing, Informatics, Decision Systems Engineering at Arizona State University. She also has a courtesy appointment of an Associate Professor of Radiology with Mayo Clinic, College of Medicine. She received her Ph.D. in Industrial Engineering from the University of Iowa in 2001. Her current research interests include: health and imaging informatics, swarm intelligence, distributed decision support. Professor Wu has articles published about 90 journal articles in journals such as IIE Transactions, IEEE Transactions on Evolutionary Computation, IEEE Transactions on Pattern Analysis and Machine Intelligence, Information Science, Neuroimaging. She is currently serving as the editor-in-chief for IIE Transactions on healthcare systems engineering.
During the past decade, with breakthroughs in systems biology, precision medicine has emerged as a novel paradigm that has transformed healthcare. Precision medicine is an approach for disease treatment and prevention that takes into account individual variability where medical imaging is a key component. This tutorial focuses on research investigating the roles of medical imaging in cancer diagnosis and treatment planning. While the cornerstone of the imaging research is mathematical and statistical modeling, the research has to take a multidisciplinary and systematic approach due to the nature of the problem. We offer a comprehensive review and discussion on four important components that form an imaging pipeline: imaging pre-processing, imaging feature extraction, feature dimensionality reduction and classification. To illustrate the clinical relevance, our in-house developed system, the imaging Multi-Texture Disease Diagnosis System (iMT-DDS) is presented with two clinical case studies, one on breast cancer diagnosis using contrast enhanced digital mammography imaging, and the other on cholangiocarcinoma using computed tomography imaging. The future directions of the imaging research are highlighted in the end.
12:15 – 1:45pm
Environmental Policy and Decision Analysis
Chief Executive Officer, Lumina Decision Systems, Inc., Los Gatos, California, USA
Max Henrion has 30 years of experience as a researcher, educator, consultant and entrepreneur, specializing in the design and effective use of decision technologies. He is the Founder and CEO of Lumina Decision Systems, which develops and publishes decision software and provides consulting and training in decision analysis in a wide variety of applications including environment and energy, R&D management and health care. His work on decommissioning offshore oil platforms won the 2014 Decision Analysis Practice Award from the Society of Decision Professionals and the INFORMS Decision Analysis Society. He was the founding President of the Association for Uncertainty and Artificial Intelligence, a Consulting Professor at Stanford University, an Adjunct Professor at Carnegie Mellon University where he was previously Associate Professor. He has an MA in Natural Sciences from Cambridge University, Master of Design from the Royal College of Art, London, and a PhD from Carnegie Mellon.
Climate change and other environmental crises create a growing need for analysts who can design and evaluate policies that reconcile environmental and economic sustainability. I’ll explain the practical value of decision analysis, influence diagrams, treating uncertainty, multi-attribute tools, creative option design, and model-driven stakeholder conversations. I’ll illustrate the use of these methods in the successful “rigs to reefs” analysis of how to decommission California’s offshore oil platforms, which achieved policy consensus.
2 – 3:30pm
Location Models for Emergency Service Applications
Department of Electrical Engineering, Ponticia Universidad Catolica de Chile, Santiago, Chile
Vladimir Marianov is a Professor at the Department of Electrical Engineering of the Pontificia Universidad Católica de Chile. He earned a M.S.E. and a Ph. D. degrees from The Johns Hopkins University in 1987 and 1989, respectively, as well as an Electrical Engineering degree from Universidad de Chile in 1978. His teaching and research interests are in the areas of facility location and network design, with applications in communications networks, nature reserve selection, location of public services, location under congestion and location of competitive facilities. He has published over 70 research papers in numerous leading journals such as Computers and Operational Research and European Journal of Operational Research. He has co-edited two books on Facility Location Foundations and Applications, and is also author of several book chapters, on subjects including Location of Emergency Systems, Location in the Public Sector, Location of Jails, and Competitive Location.
The term “emergency services” commonly refers to medical, firefighting or police services, although emergency repair can also be included in this category. All these services can be found in both public and private sectors. The design of police services requires mainly districting, dispatching, routing, and scheduling of police beats. Medical and firefighting services, on the other hand, mainly involve locating infrastructure and finding an initial or “idle” site where vehicles wait for an emergency to occur. Thus, the design of medical and firefighting systems benefit the most from the use of location models, and we concentrate in these two types of services. We do not perform a thorough review of the literature related to the application of location models to these services. Rather, we follow some of the successive model innovations that have addressed the main issues and added increasing degrees of reality to the modeling, with a didactic goal. We thus apologize to colleagues who have made significant contributions not included in this chapter.
4:30 – 6pm
Competition in Multi-Echelon Systems
Graduate School of Business, Columbia University, New York, New York 10027
Awi Federgruen is the Charles E. Exley Professor of Management of the Decision, Risk and Operations (DRO) Division of the Graduate School of Business at Columbia University where he served as Senior Vice Dean from 1997-2002 and as the Chair of the DRO division for many years. He obtained his doctorate in Operations Research at the University of Amsterdam in the Netherlands. He is a world renowned expert in the development and implementation of planning models for supply chain management and logistical systems. Much of his recent work focuses on competition, coordination and contracting within supply and service chains. He is the recipient of the 2004 Distinguished Fellowship Award by the Manufacturing, Service and Operations Management society, and was elected a presidential Fellow of the INFORMS society. He is currently the Editor-in-Chief of Naval Research Logistics, and a former Departmental Editor for the Department of Manufacturing, Service and Operations of Management Science. He also served as Associate Editors for several flagship journals of his profession.
Rotman School of Management, University of Toronto, Toronto, Ontario, Canada M5S 3E6
Ming Hu is a Professor of Operations Management at Rotman School of Management, University of Toronto. He received a master’s degree in Applied Mathematics from Brown University in 2003, and a Ph.D. in Operations Research from Columbia University in 2009. His research explores strategic interactions among firms and between firms and consumers, in the context of revenue management, supply chain management and service operations management. Most recently, he focuses on operations management in the context of social buying, crowdfunding, crowdsourcing, and two-sided markets, with the goal to exploit operational decisions to the benefit of the society. He is the recipient of 2016 Wickham Skinner Early-Career Research Accomplishments Award by Production and Operations Management Society (POMS) and 2017 Management Science Best Paper Award in Operations Management.
Around the start of this new millennium, scholars in the operations management/operations research field started to make important contributions to the study of price competition models. In this tutorial, we review these contributions, and partition them into five broad areas. Most of the tutorial is devoted to the last and most recent category: price competition models for multi-echelon supply chains with an arbitrary number of competing firms and products at each echelon.