ORMS Today December 2016 by Lionheart Publishing Inc.

AWARDS GALORE: INFORMS salutes its brightest stars

December 2016

Volume 43 • Number 6 ormstoday.informs.org

Inside: Why the polls were wrong The presidential election that confounded everyone

O.R. at Google Tech giant a ‘candy store’ for practitioners

Sound data science Avoiding most pernicious prediction pitfall

Balancing risk with probability PG&E case study: Emerging discipline of probability management provides overall risk snapshot to assess tradeoffs between safety, reliability and cost

Moving ahead with membership Interview with incoming INFORMS President Brian Denton

Contents December 2016 | Volume 43, No. 6 | ormstoday.informs.org

On the Cover Risky balancing act Case study: California power company PG&E turns to emerging discipline of probability management to balance risk and probability. Image © Tsung-Lin Wu | 123rf.com

18 de partm e nt s F e at ure s 18

Roundtable profile: O.R. at Google

Rolling up operational risk at PG&E

The election that confounded everybody

By Brian Thomas Eck and Amber Richter Advanced analytics permeates work at Google, making the technology giant a ‘candy store for O.R. practitioners.’

By Jordan Alen, Christine Cowsert Chapman, Melissa Kirmse, Farshad Miraftab and Sam Savage Probability management provides overall risk snapshot to assess tradeoffs between safety, reliability and cost.

2 | ORMS Today

6 8 10 12 14 16 58 59 64

Inside Story President’s Desk INFORMS in the News Issues in Education PuzzlOR INFORMS Initiatives Industry News Classifieds ORacle

By Douglas A. Samuelson Why the polls were wrong: The “13 Keys” model outperformed its creator during an unpredictable presidential campaign.

December 2016

ormstoday.informs.org

Welcome to Analytic Solver ® Cloud-based Optimization Modeling that Integrates with Excel

Everything in Predictive and Prescriptive Analytics Everywhere You Want, from Concept to Deployment. The Analytic Solver® suite makes the world’s best optimization software available in your web browser (cloud-based software as a service), and in Microsoft Excel. And you can easily create models in our RASON® language for your server, web and mobile apps.

Linear Programming to Stochastic Optimization. It’s all point-and-click: Fast, large-scale linear, quadratic and mixed-integer programming, conic, nonlinear, non-smooth and global optimization. Easily incorporate uncertainty and solve with simulation optimization, stochastic programming, and robust optimization.

And it’s a full-power, point-and-click tool for forecasting, data mining and text mining, from time series methods to classification and regression trees, neural networks and association rules – complete with access to SQL databases, Apache Spark Big Data clusters, and text document folders.

Find Out More, Start Your Free Trial Now. In your browser, in Excel, or in Visual Studio, Analytic Solver comes with everything you need: Wizards, Help, User Guides, 90 examples, even online training courses. Visit www.solver.com to learn more or ask questions, and visit analyticsolver.com to register and start a free trial – in the cloud, on your desktop, or both!

Monte Carlo Simulation, Data and Text Mining. Analytic Solver is also a full-power, point-and-click tool for Monte Carlo simulation and decision analysis, with a Distribution Wizard, 50 distributions, 50 statistics and risk measures, and a wide array of charts and graphs.

Tel 775 831 0300 • Fax 775 831 0314 • info@solver.com

December 2016 | Volume 43, No. 6 | ormstoday.informs.org

INFORMS Board of Directors

President Edward H. Kaplan, Yale University President-Elect Brian Denton, University of Michigan

Past President L. Robin Keller, University of California, Irvine

32 32

Vice President-Meetings Ronald G. Askin, Arizona State University Vice President-Publications Jonathan F. Bard, University of Texas-Austin

Vice President- Esma Gel, Arizona State University Sections and Societies

Vice President- Marco Lüebbecke, Information Technology RWTH Aachen University

F e at ure s

In search of sound data science By Eric Siegel Predictive analytics: Avoiding the most pernicious prediction pitfall – the potential and danger of automating science.

Secretary Pinar Keskinocak, Georgia Tech

Treasurer Sheldon N. Jacobson, University of Illinois

Vice President- Jonathan Owen, CAP, General Motors Practice Activities Vice President- Grace Lin, International Activities Institute for Information Industry

Vice President-Membership Susan E. Martonosi, Professional Recognition Harvey Mudd College Vice President-Education Jill Hardin Wilson, Northwestern University Vice President-Marketing, Laura Albert McLay, Communications and Outreach University of Wisconsin-Madison Vice President-Chapters/Fora Michael Johnson, University of Massachusetts-Boston

Editors of Other INFORMS Publications

Moving forward with membership help By Peter Horner Q&A: 2017 INFORMS President Brian Denton discusses goals, aspirations and member-inspired ideas, initiatives.

Decision Analysis Rakesh K. Sarin, University of California, Los Angeles

I NFORMS Journal on Computing David Woodruff, University of California, Davis

n ews 45 Analytics Conference

51 Dissertation Award

45 Board Nominations

51 Nicholson Student Paper

INFORMS Online Kevin Geraghty, 360i INFORMS Transactions Jeroen Belien, KU Leuven on Education

Interfaces Srinivas Bollapragada, General Electric Global Research Center Management Science Teck-Hua Ho, National University of Singapore Office of the Deputy President (Research and Technology)

Editor’s Cut Anne G. Robinson, Verizon Wireless

Information Systems Research Ritu Agarwal, University of Maryland

Manufacturing & Service Christopher S. Tang, Operations Management University of California, Los Angeles

Marketing Science K. Sudhir, Yale University

Mathematics of Operations J. G. “Jim” Dai, Cornell University Research

Operations Research Stefanos Zenios, Stanford University

Organization Science Zur Shapira, New York University

46 INFORMS Fellows

52 Doing Good O.R. Prize

47 President’s Award

52 Volunteer Service Award

Service Science Paul P. Maglio, University of California, Merced

47 Impact Prize

53 People

Strategy Science Daniel A. Levinthal, Wharton School, University of Pennsylvania

48 Kimball Medal

53 Tribute: Marius Solomon

Transportation Science Martin Savelsbergh, Georgia Institute of Technology

49 Expository Writing Award

54 Photos from Nashville

49 Undergraduate O.R. Prize

56 Snapshot Survey

50 von Neumann Prize

56 Meetings

51 Teaching OR/MS Prize

4 | ORMS Today

December 2016

Tutorials in Operations J. Cole Smith, University of Florida Research

INFORMS Office • Phone: 1-800-4INFORMS

Executive Director Melissa Moore

Headquarters

INFORMS (Maryland) 5521 Research Park Dr., Suite 200 Catonsville, MD 21228 USA Tel.: 443.757.3500 Fax: 443.757.3515 E-mail: informs@informs.org

ormstoday.informs.org

Welcome to Analytic Solver ® Cloud-based Simulation Modeling that Integrates with Excel

Everything in Predictive and Prescriptive Analytics Everywhere You Want, from Concept to Deployment. The Analytic Solver® suite makes the fastest Monte Carlo simulation and risk analysis software available in your web browser (cloud-based software as a service), and in Microsoft Excel. And you can easily create models in our RASON® language for server, web and mobile apps.

Comprehensive Risk and Decision Analysis Tools. Use a point-and-click Distribution Wizard, 50 probability distributions, automatic distribution fitting, compound distributions, rank-order correlation and three types of copulas; 50 statistics, risk measures and Six Sigma functions; easy multiple parameterized simulations, decision trees, and a wide array of charts and graphs.

nonlinear optimization, simulation optimization, stochastic programming and robust optimization. And it’s a full-power tool for forecasting, data mining and text mining, from time series methods to classification and regression trees, neural networks and more, with access to SQL databases and Spark Big Data clusters.

Optimization, Forecasting, Data and Text Mining. Analytic Solver is also a full-power, point-and-click tool for conventional and stochastic optimization, with powerful linear and mixed-integer programming,

Tel 775 831 0300 • Fax 775 831 0314 • info@solver.com

Inside Story

Peter Horner, editor

peter.horner@mail.informs.org

Politics vs. pollsters

OR/MS Today Advertising and Editorial Office Send all advertising submissions for OR/MS Today to: Lionheart Publishing Inc. 506 Roswell Street, Suite 220, Marietta, GA 30060 USA Tel.: 888.303.5639 • Fax: 770.432.6969

President

On the morning of Nov. 8, Election Day, one analytics-oriented website indicated Hillary Clinton had a 99 percent chance of garnering the 270 electoral votes needed to win the U.S. presidential election. A few days earlier, statistics guru Nate Silver of “FiveThirtyEight” fame gave Clinton a 75 percent chance of winning the election. Meanwhile, in the weeks leading up to the election, virtually every credible “scientific” poll showed little or no viable path through the Electoral College to the White House for Donald Trump. How could so many pollsters and political pundits get it so wrong? That’s a question a lot of people – with the possible exception of quantitative historian Allan Lichtman – have been asking since the results of Nov. 8 stunned the world. In mid-September, Lichtman, a history professor at American University in Washington, D.C., publicly predicted a Trump triumph despite the fact that Clinton held a substantial lead at the time in most national polls. So then the question becomes, how did Lichtman get it right? The answer: by accident. As chronicled in his book (“Predicting the Next President: The Keys to the White House”) and in the mainstream media, as well as in the pages of OR/MS Today for the last 20 years, Lichtman bases his presidential predictions on a “13 Keys” model – a ser ies of 13 true/false statements concerned for the most part with the incumbent administration. If six or more of the statements are false, the incumbent party candidate, in this case Hillary Clinton, loses. The Keys model focuses on incumbent governance rather than the perceived strengths or weaknesses of the current candidates and their campaigns. So where did Lichtman go astray? Well, his model comes with a caveat: It is designed to predict the winner 6 | ORMS Today

December 2016

of the national vote, not the Electoral College vote. Until this year, the model correctly predicted the winner of the national vote in every presidential election since 1984, including Al Gore’s “win” in 2000. At the time of this writing, however, Clinton held about a two million lead in the 2016 national vote. So, due to misinterpreting one of the keys, Lichtman apparently got the national vote wrong but correctly predicted the next president. D o u g S a mu e l s o n i n t ro d u c e d Lichtman to operations research and OR/ MS Today readers in 1996 (“Unlocking the Door to the White House,”), and Samuelson has been writing about the “13 Keys” every four years since. A prolific contributor to OR/MS Today on a variety of topics, Samuelson is perhaps best known to readers as the author of the “ORacle”– this issue marks the completion of the 30th year of the column. In the July-August 2016 issue of Analytics magazine, Samuelson gave a preview of this year’s election based on the 13 Keys model. At the time, three of the keys had yet to be determined, so Lichtman held off on making a prediction. For this issue of OR/MS Today, Samuelson again touches base with Lichtman for a post mortem on the election, predictions and lessons learned during this most unusual and unpredictable of presidential campaigns. For more on the story, including the advent of political “tribalism” and a nod to Marshall McLuhan who famously stated that the “medium is the message” and who in 1964 predicted the multimedia mess of an election we witnessed this year, see “The election that confounded everybody” (page 38). Clearly, political pollsters, pundits and prognosticators – and those who do the analytics and number crunching in support of all of the above – have a lot to learn between now and 2020. ORMS

John Llewellyn, ext. 209 john.llewellyn@mail.informs.org

Editor Peter R. Horner peter.horner@mail.informs.org Tel.: 770.587.3172

Assistant Editor Donna Brooks

Contributing writers/editors Douglas Samuelson, Matt Drake, John Toczek

Art Director Alan Brubaker, ext. 218 alan.brubaker@mail.informs.org

Online Projects Manager Patton McGinley, ext. 214 patton.mcginley@mail.informs.org

Assistant Online Projects Manager Leslie Proctor, ext. 228 leslie.proctor@mail.informs.org

Advertising Sales Managers Sharon Baker sharon.baker@mail.informs.org Tel.: 813-852-9942 Aileen Kronke, ext. 212 aileen@lionhrtpub.com

Reprints Kelly Millwood, ext. 215 kelly.millwood@mail.informs.org

OR/MS Today Committee James Cochran, chairman

INFORMS Online http://www.informs.org

Lionheart Publishing Online http://www.orms-today.org OR/MS Today (ISSN 1085-1038) is published bimonthly by the Institute for Operations Research and the Management Sciences (INFORMS). Canada Post International Publications Mail (Canadian Distribution) Sales Agreement No. 1220047. Deadlines for contributions: Manuscripts and news items should arrive no later than six weeks prior to the first day of the month of publication. Address correspondence to: Editor, OR/MS Today, 506 Roswell Street, Suite 220, Marietta, GA 30060. The opinions expressed in OR/MS Today are those of the authors, and do not necessarily reflect the opinions of INFORMS, its officers, Lionheart Publishing Inc. or the editorial staff of OR/MS Today. Membership subscriptions for OR/MS Today are included in annual dues. INFORMS offers non-member subscriptions to institutions, the rate is $62 USA, $79 Canada & Mexico and $85 all other countries. Single copies can be purchased for $10.50 plus postage. Periodicals postage paid at Catonsville, MD, and additional mailing offices. Printed in the United States of America. POSTMASTER: Send address changes to OR/MS Today, INFORMS-Maryland Office, 5521 Research Park Dr., Suite 200, Catonsville, MD 21228. OR/MS Today copyright ©2016 by the Institute for Operations Research and the Management Sciences. All rights reserved.

ormstoday.informs.org

Welcome to Analytic Solver ® Cloud-based Data and Text Mining that Integrates with Excel

Everything in Predictive and Prescriptive Analytics Everywhere You Want, from Concept to Deployment. The Analytic Solver® suite makes powerful forecasting, data mining and text mining software available in your web browser (cloud-based software as a service), and in Microsoft Excel. And you can easily create models in our RASON® language for server, web and mobile apps.

Full-Power Data Mining and Predictive Analytics. It’s all point-and-click: Text mining, latent semantic analysis, feature selection, principal components and clustering; exponential smoothing and ARIMA for forecasting; multiple regression, logistic regression, k-nearest neighbors, discriminant analysis, naïve Bayes, and ensembles of trees and neural networks for prediction; and association rules for affinity analysis.

distributions, 50 statistics and risk measures, rankorder and copula correlation, distribution fitting, and charts and graphs. And it has full-power, point-and-click optimization, with large-scale linear and mixed-integer programming, nonlinear and simulation optimization, stochastic programming and robust optimization.

Simulation/Risk Analysis, Powerful Optimization. Analytic Solver is also a full-power, point-and-click tool for Monte Carlo simulation and risk analysis, with 50

Tel 775 831 0300 • Fax 775 831 0314 • info@solver.com

President’s Desk

‘Me m

b e r-in- C h ie f Me m o’

Ed Kaplan

INFORMS President president@informs.org

Celebrating the operations research mindset At a national society meeting, the president of INFORMS stated: “ We n e e d t o d e v e l o p n e w methodology and to adapt old; we need to generalize our basic theoretical techniques and to broaden their range of application.” Who said this and when? The speaker was Phil Morse, the first president of the Operations Research Society of America (ORSA), and the occasion was the ORSA Annual Meeting in May 1953. Morse’s basic prescription – keep building up our theory; keep expanding our applications – works just as well today as a guide for the future of operations research. Others have proffered up their own views of the future of operations research since Morse first looked into the crystal ball. Some of these are devastating in their pessimism. In 1979, Russell Ackoff wrote, “The life of O.R. has been a short one – it was born late in the 1930s – by the mid-1960s most O.R. courses were given by academics who never practiced it, depriving O.R. of its unique incompetence.” Ackoff argued that we “... should want to help create a world in which the capabilities of O.R. are considerably extended but in which the need for O.R. is diminished.” This does not sound like a recipe for growing a discipline. With a healthy 12,000+ membership, INFORMS has happily not followed Ackoff ’s advice. Others have provided more optimistic views. Ten year s post-Ackoff , the irrepressible Alexander Rinnooy Kan wrote that,“The future of O.R. is bright – if there is anything worrying about the state of O.R., it is that our discipline seems to spend such an inordinate amount of time and effort worrying about itself.” 8 | ORMS Today

December 2016

Operations researchers are the masters of structuring messy situations into problems amenable to analysis.

Who can’t relate to that? What should our name be: Operations research? Management science? Decision sciences? Analytics? Calcuholics? This annoying issue has been around for a long time. In a 1952 article in the Journal of Applied Physics, Phil Morse wrote: “Personally, I would prefer to forget about definitions and get on with the work. After all, who cares what it’s called, as long as it’s useful and is used?” 1952!! Our methodological focus might seem odd to outsiders who do not appreciate the history of our field.Why are our main mathematical tools rooted in optimization and stochastic processes? Why not number theory and topology? The answer is that our discipline is rooted in the scientific study of operations, those tasks and processes that represent how organizations “get things done.” Such study is meant to improve decisions, which explains why optimization is so important, while randomness and uncertainty abound in operational processes, making applied probability expertise essential. Advances in theory have intrinsic value, like art or music, beyond that offered from use in future applications. And while methods and tools evolve over time, our basic approach of using models to better understand systems and improve their performance has stood the test of time. Going back to Morse for a moment, he was really excited about using analogue devices for teaching O.R. One exciting educational application went like this: “A radioactive source and two Geiger counters provide two purely random sequences of pulses, which may

be varied in mean rate merely by changing the distance of the counter from the source. For example, one counter can represent arrivals in a queue, and the other can represent the service operation that removes the individual from the waiting line; an electronic counter can then indicate the instantaneous length of the queue.” Talk about glowing customers! O p e r a t i o n s r e s e a r c h , u n l i ke economics (or physics for that matter), does not possess a “world view” – we have no underlying holistic theory for how the world works. The natural unit of interest in O.R. is “the problem.” It shows in how we label things – the diet problem, traveling salesman problem, stochastic queue median problem, etc. – and it shows in how we decompose more complicated situations into something we can study, model, understand and perhaps improve. But operations research has a mindset. Operations researchers are the masters of structuring messy situations into problems amenable to analysis. Operational science includes seeing or characterizing phenomena of all sorts as operations. Modeling science (or perhaps modeling art) calls upon our creativity to create new models for such operations. These are key O.R. skills, and they capture what many INFORMS members really do. Sometimes, immersion in a particular problem domain leads operations researchers to become subject matter experts. Thus, Jon Caulkins has figured ormstoday.informs.org

Operations research is a terrific, wonderful area of endeavor of which you should all be proud. out the optimal price for cocaine, Larry Wein and Jerry Brown have secured the homeland, Margaret Brandeau optimally allocates public health resources, Dimitris Bertsimas will ensure that optimization is robust, Arnie Barnett can tell you when your plane will crash, Garrett van Ryzin will get you a seat on that plane at a lower price, Linda Green and Carri Chan will divert an ambulance to get you to the hospital, and Ralph Keeney can explain why it’s all your fault. O.R. is not an add-on to such expertise; rather O.R. was crucial in establishing this expertise in the first place. So, with mindset and domain expertise in place, all of us can rally to our core purpose of advancing our

science and practice. All of us can contr ibute to helping decision-makers use our technologies, and enable organizations to institutionalize our approaches in their own decision processes. And, all of us can, at least in some small way, use our expertise to help make the world a better place. Phil Morse had it right 60 years ago. We need to develop new methodology and to adapt old; we need to generalize our basic theoretical techniques and broaden their range of application. Some final thoughts from your departing member-in-chief: We can have a lot of fun doing these things while celebrating how our field has helped us lead more meaningful lives. Operations research is a terrific, wonderful area of endeavor of which you should all be proud. Keep doing stuff! ORMS

INFORMS President Ed Kaplan (right) was busy “doing stuff” at the 2016 INFORMS Annual Meeting, in this case welcoming Tamás Terlaky (left) as a member of the 2016 class of Fellows. More coverage of the Fellows and other awards begins on page 46.

Which organization has the best O.R. department in the world?

prize

call for

nominations

DEADLINE TO APPLY IS DECEMBER 15, 2016

The Institute for Operations Research and the Management Sciences annually awards the INFORMS Prize for effective integration of Operations Research/Management Science (OR/MS) and advanced analytics into organizational decision making. The award is given to an organization that has repeatedly applied the principles of OR/MS and advanced analytics in pioneering, varied, novel, and lasting ways.

log onto: www.informs.org/informsprize for more information or contact Tarun Mohan Lal, Principal, Advanced Analytics Clinic Operations Mayo Clinic voice: +1 507-266-9842 email: mohanlal.tarun@mayo.edu

2017

Tell us why the title should be yours. December 2016

ORMS Today

| 9

INFORMS in the News

Compiled by Ashley Kilgore

Cool factor, deadlines, soup science and more INFORMS members, initiatives and journals continue to make news on a wide range of topics in a variety of forums. Following are recent examples of “INFORMS in the News”: Reducing the costs of chronic illness A forthcoming study in the INFORMS journal Marketing Science by Jian Ni, INFORMS member and professor at Johns Hopkins Carey Business School, explores how to reduce healthcare expenditures by guiding patients to a more appropriate level of healthcare. - Changing Business, Johns Hopkins Carey Business School, Sept. 16

Measuring the cool factor In an upcoming study in the INFORMS journal Management Science, INFORMS member and professor at the Johns Hopkins Carey Business School, Ruxian Wang, has developed a method to measure the appeal or “cool factor” of commercial products and how this impacts consumer spending. - Changing Business, Johns Hopkins Carey Business School, Sept. 16

O.R. master’s degree has third highest salary potential According to PayScale’s “2016–2017 College Salary Report,” an operations research master’s degree ranks third behind only computer science and engineering, and nurse anesthesia for the highest salary potential. - Forbes, Sept. 22

Strengthening the ties between students and industry Executive director of INFORMS, Melissa Moore, presented Carnegie 10 | ORMS Today

December 2016

Mellon University’s H. John Heinz III College with the UPS George D. Smith Prize.This prize is awarded in recognition of the school’s encouragement of stronger ties between industry and students studying operations research and analytics, in particular, for student projects relating to healthcare and real estate. - Pittsburgh Business Times, Sept. 26

School of Business, Professors Yili Hong and Zhongju Zhang, in conjunction with research associate Ziru Li, have conducted a study that explores the positive impact of ride-hailing services, including reduced traffic congestion, travel time and gas expenses. Their study was presented at the INFORMS Annual Meeting in November. - Science Blog, Sept. 30

Exploring the CAP exam An editorial by Arnie Greenland, Ph.D., CAP, professor of the Practice at the Robert H. Smith School of Business at the University of Maryland, College Park, explores the foundation of the CAP certification program, its growing recognition among employers, and provides a sneak peek at what to expect when taking the exam. - KDnuggets, Oct. 12

Are strict deadlines impacting your work quality? A new study in Management Science, led by Syracuse University professor Natarajan Balasubramanian, found that when managers implement strict deadlines, their workers tend to complete their tasks at the last minute, which often leads to lower quality outcomes. - Eenadu India, Sept. 28

Betting to lose helps you win From spor ts teams to political candidates, when your side loses it can be painful. An upcoming study in the INFORMS journal Management Science finds that this pain can be lessened if you bet against your team or candidate, though feelings of disloyalty may prevent many from doing so. - The New York Times, Oct. 14

Increase both online and in-store purchases with online display ads A recent study in the INFORMS journal Marketing Science conducted by Yahoo! Research in partnership with a nationwide retailer explored the effects of online display advertising on both online and in-store purchases. The study authors found statistically significant evidence that the retailer ads increased sales 3.6 percent relative to the control group.

An O.R. career means high pay with low stress Based on data pulled from the Occupational Information Network (O*NET) – a U.S. Department of Labor database that compiles detailed infor mation on hundreds of jobs – and the U.S. Bureau of Labor Statistics website, operations research analysts are among a list of 26 jobs that combine high pay with low stress.

- Consumer Goods Technology, Sept. 30

- Independent, Oct. 19

Exploring the ‘Uber effect’ Two INFORMS members at the Arizona State University W.P. Carey

Enhancing oil system safety I N F O R M S m e m b e r Je n n i f e r Pazour, assistant professor of industrial ormstoday.informs.org

and systems engineering at Rensselaer Polytechnic Institute, has been named a recipient of a 2016 Gulf Research P rog r a m E a r l y - C a re e r R e s e a rc h Fellowship, a prog ram that funds activities to enhance oil system safety and the protection of human health and the environment in the Gulf of Mexico and United States.

O.R. analyst ranks among top paying jobs for women Operations research analyst ranks at number 10 on a list of the 17 highest paying jobs for women. - GoBankingRates.com, Oct. 20

The science of soup You may not realize it, but a significant amount of science and technology is behind every can of soup. INFORMS member Joseph Byrum, senior R&D and strategic marketing executive with Syngenta, explores the extensive research that goes into plant genetics and breeding to canning to create a “simple” can of tomato soup.

The pub crawl of a lifetime Planning a pub crawl and need to know the shortest distance between each establishment in which you hope to imbibe? INFORMS member William Cook, professor with the University of Waterloo, Canada, has done just that on a much grander scale than could be accomplished in one night of festivities. Using the “traveling salesman problem” approach, Cook plotted the coordinates of 24,727 pubs in the U.K. to ascertain the shortest possible route between them all.

- Fast Company, Oct. 20

– The Guardian, Oct. 21

- NewsWise, Oct. 19

The benefits of working outside your field After dedicating a significant number of years and money to earning a university degree, what motivates people to voluntar ily work in jobs outside their field? In a study selected for publication in the INFORMS journal Organization Science, study author Briana Sell Stenard, an INFORMS member and assistant professor of management and entrepreneurship at the Stetson School of Business and Economics at Mercer University, explored the factors that contribute to this including loss of interest, better pay and working conditions, higher positions and increased flexibility. – Harvard Business Review, Oct. 31

For links to all of the articles mentioned above, see http://bit.ly/2h2qmBU. ORMS Ashley Kilgore (akilgore@informs.org) is the public relations manager at INFORMS.

! PYNE O I C L

R ON OUailable bers! Y m S av

ES Rials MS Me C C tO R

u 6 T r INFO 1 0 2 fo

Optimization Challenges in Complex, Networked, and Risky Systems 2016 Optimization Challenges in Complex, Networked, and Risky Systems Aparna Gupta and Agostino Capponi, Tutorials Co-Chairs and Volume Editors J. Cole Smith, Series Editor 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 Meeting.

INFORMS 2016 edition of the TutORials in Operations Research series will be available online to registrants of the 2016 INFORMS Annual Meeting on November 12, 2016. The fifteen chapters in this year’s volume highlight the tutorial theme of Optimization Challenges in Complex, Networked, and Risky Systems. The chapters are written by a diverse array of experts working across a variety of institutes. Their research covers a range of exciting topics, including multiobjective optimization, optimization under uncertainty, big data analytics, project management, and risk modeling and optimization. The chapters focus on many compelling applications for which this analysis is useful, including those arising in finance, healthcare, and energy systems.

Access the 2016 TutORials at:

http://pubsonline.informs.org/series/educ

Aparna Gupta and Agostino Capponi, Tutorials Co-Chairs and Volume Editors

J. Cole Smith, Series Editor

Optimization Challenges in Complex, Networked, and Risky Systems

2016

Aparna Gupta and Agostino Capponi, TutORials Co-Chairs and Volume Editors

www.informs.org

Nashville 2016 Presented at the INFORMS Annual Meeting, November 13– 16, 2016

December 2016

ORMS Today

| 11

Issues in Education

By James J. Cochran jcochran@cba.ua.edu

You, INFORMS and OR/MS education Are you familiar with the wide array of INFORMS education-oriented initiatives and activities? I spoke about this with dozens of members at the recent INFORMS conference in Nashville and was I surprised at how many knew little or nothing about these efforts. I was able to (somewhat) rectify this problem with several of these colleagues one-on-one, but taking this approach to educating each member on INFORMS’ various education initiatives, while gratifying and worthwhile, would be extremely inefficient. I volunteered to contribute to the OR/MS Today “Issues in Education” column as a means of more efficiently addressing this issue. Let’s start with the initiative around which the greatest number of members is directly involved – INFORM-ED. INFORM-ED (https://www.informs. org/Community/INFORM-ED) is INFORMS’ education forum. What is a forum? INFORMS’ structure of communities includes four broad classes: chapters, sections, societies and fora (or forums, if you prefer). A chapter is a geography-based group of INFORMS members; INFORMS has dozens of chapters throughout the United States and a few outside of the U.S. (such as the Korean and Polish chapters). A section is a group of INFORMS members who share some relatively narrow professional interest; INFORMS has many sections including Data Mining, Telecommunications and SpORts. A forum is a group of colleagues with a common professional interest that extends beyond disciplines and geography; INFORMS five fora are the Association of Chairs of O.R. Departments (ACORD), the Junior Faculty Interest Group (JFIG), the Minority Issues Forum (MIF),Women in OR/MS (WORMS) and the Forum on Education (INFORM-ED). A society 12 | ORMS Today

December 2016

I a s ke d Pa l a n i a p p a K r i s h n a n , associate professor with the University of Delaware’s Applied Economics and Statistics Department and chair of the 2016 Case and Teaching Mater ials Competition, to share his thoughts on the competition. His response was:

This incredibly productive forum provides many important services to the OR/MS community and INFORMS members.

is basically an overgrown section – a large group of INFORMS members with a common, relatively narrow professional interest; INFORMS has many societies including Optimization, Analytics and Military Applications. I t i s e a s y t o u n d e r s t a n d w hy INFORM-ED is a forum rather than a section or society – OR/MS education is (or should be) of interest to all INFORMS members! Whether you work in academia, private industry or government, you have a vested interest in attracting bright young people into our discipline and providing them with the best possible OR/MS education experience. For this reason, it is difficult for me to understand why less than 1 percent of all INFORMS regular members belong to INFORM-ED. This incredibly productive forum provides many important services to the OR/ MS community and INFORMS members. It publishes a newsletter that provides its members with information on education initiatives, publishing outlets, workshops, software and other topics of interest. It sponsors this “Issues in Education” column in OR/MS Today. It organizes a deep and diverse track of informative sessions for the INFORMS Annual Meeting. It established and now organizes and sponsors the annual Teaching Effectiveness Colloquium that has been held annually prior to the annual INFORMS conference every year since 1999. And it established and sponsors the prestigious Case and Teaching Materials Competition that has been held during the INFORMS Annual Meeting every year since 2000. As I renew my INFORMS membership for 2017, I note that annual membership in the Forum on Education currently costs a regular member only $10! I do not know of too many bigger bargains.

“The INFORMS Case Competition is a great place to showcase one’s original teaching cases, and it is a terrific learning platform for young assistant professors who are trying to build a teaching portfolio. They can get good feedback about their cases and network with likeminded colleagues, and all faculty can find good material to use in their classrooms. Organizations that would like to work with faculty to develop teaching cases around their real experiences can also find potential partners for these endeavors.” I asked Mihai Banciu, associate professor of Operations and Decision Sciences in the Bucknell University School of Management and chair of the 2016 Teaching Effectiveness Colloquium, to share his thoughts on OR/MS education, INFORM-ED and the Teaching Effectiveness Colloquium. Mihai’s response follows: “To me, participating in TEC is important for two reasons: First, I don’t really think that one can truly separate research from teaching, and so I argue that by becoming a better teacher one does in fact become a better researcher and vice-versa. That is a “selfish” reason to attend TEC. Second, I think that we owe it to ourselves to train the next generation of people who will move the discipline of OR/MS forward. This is the “altruistic” reason to attend TEC.” Mihai added that he was reminded of a Richard Feynman quote from ormstoday.informs.org

“Surely you’re joking, Mr. Feynman.” The full quote with context is available at http://www.math.utah.edu/~yplee/ teaching/feynman.html, but here’s a portion of it: “If you’re teaching a class, you can think about the elementary things that you know very well. These things are kind of fun and delightful. It doesn’t do any harm to think them over again. Is there a better way to present them? Are there any new problems associated with them? Are there any new thoughts you can make about them? The elementary things are easy to think about; if you can’t think of a new thought, no harm done; what you thought about it before is good enough for the class. If you do think of something new, you’re rather pleased that you have a new way of looking at it. “The questions asked by students are often the source of new research ideas. They often ask profound questions that I’ve thought about at times and then given up on, so to speak, for a while. It wouldn’t do me any harm to think about them again and see if I can go any further now. The students may not be able to see the research question I want to answer, or the subtleties I want to consider, but they remind me of a research problem by asking questions in the neighborhood of that problem. It’s not so easy to remind yourself of these things. “So I find that teaching and the students keep life going, and I would never accept any position in which somebody has invented a happy situation for me where I don’t have to teach. Never.” I n a re c e n t c o r re s p o n d e n c e, INFORMS Member-in-Chief Ed Kaplan also shared his thoughts on the role education and INFORM-ED play in our institute. Ed responded in the following way:

“Few things are more important to the health of our f ield – and to INFOR MS – than continuing to attract and support talented students, and providing them with the best educational expe r ien ce possible. INFOR M-ED is a critical part of our activities in this regard. For example, INFORM-ED INFORM-ED provides many important services to the OR/MS supports the publication community and INFORMS members. of the open-access Image © dotshock | 123rf.com journal IN FOR MS Transactions on Education, Competition. Register for the 2017 which provides all manner of Teaching Effectiveness Colloquium pedagogical discussions and classroom (a full day of short workshops on discussions oriented towards teaching strateg ies for improving OR/MS operations research and analytics. education – completely underwritten Another example is the INFORMS by INFORMS). Write an “Issues in Case Competition, which promotes Education” column for OR/MS Today. the development of new teaching If you are working in pr ivate cases and associated instructional industry or for a government agency, m a t e r i a l s . T h e Te a c h i n g identify applications of OR/MS within Effectiveness Colloquium offers your organization – successes, failures new instructors the opportunity to and ever ything in-between – and learn from experts regarding effective work with an instructor to turn these organization and presentation. applications into teaching cases that can then be published and made freely “It is so critical for a small field available by INFORMS Transactions like ours to intellectually reproduce. on Education (this is something our D o i n g so r e q u i r e s a t t ra c t i n g colleagues in developing nations greatly excellent students to be sure, but appreciate). Get involved in some way – ensuring that such students are the continued growth of our discipline turned on is important for retention and our institute depends on it.ORMS whether said students turn to the James J. Cochran (jcochran@cba.ua.edu) is academic or practice worlds (or a professor of applied statistics and RogersSpivey Faculty Fellow in the Department both). Please consider participating of Information Systems, Statistics and in INFORM-ED – it could be a Management Science, Culverhouse College turn-on for you too!” I encourage you to join INFORM-ED when you renew your INFORMS membership for 2017. But don’t stop there. Volunteer to review ar ticles for INFORMS Transactions on Education or serve as a judge for the Case and Teaching Mater ials Competition. Submit an article to INFORMS Transactions on Education or a case to the Case and Teaching Materials

of Commerce and Business Administration, University of Alabama-Tuscaloosa. A longtime, active member of INFORMS and INFORMS-ED, Professor Cochran received the 2008 INFORMS Prize for the Teaching of OR/ MS Practice. In addition, he received the Mu Sigma Rho Statistical Education Award (2010), was named a Fellow of the American Statistical Association (2011), and was a recipient of the American Statistical Association’s Founders Award (2014) and the Karl E. Peace Award for outstanding statistical contributions for the betterment of society (2015).

December 2016

ORMS Today

| 13

PuzzlOR

John Toczek

puzzlor@gmail.com

Ko jad in o

vic

3r f.c o

In the market for a spaceship?

Question: Which is the most undervalued ship? Send your answer to puzzlor@gmail.com by Feb. 15, 2017. The winner, chosen randomly from correct answers, will receive a $25 Amazon Gift Card. Past questions and answers can be found at puzzlor.com. ORMS John Toczek earned his BSc. in chemical engineering at Drexel University (1996) and his MSc. in operations research from Virginia Commonwealth University (2005).

ilo sh

A popular game in the Apple and Google app stores is a space adventure game where, among other things, you need to decide which spaceship to buy.There are dozens of different spaceships to choose from, all with varying capabilities and costs. Ten of these ships are shown in the accompanying table. Although there are additional ship capabilities in the game, only consider the ones shown in the table. Ship Name Cicero Cormorant Dace Hatsuyuki H’Soc N’Tirrk Razor Taipan Teneta Type 43

Cargo

Weapons

Equipment

Cost ($)

25 350 38 28 45 80 60 50 65 30

2 0 4 2 2 2 4 3 2 2

6 8 5 8 7 8 6 5 7 6

$51,975 $168,900 $235,600 $171,900 $150,000 $250,000 $294,900 $100,100 $125,400 $72,500

Table 1.

THE ODDS ARE IN YOUR FAVOR AT ANALYTICS 2017! Exhibitor and Sponsorship Opportunities at the 2017 INFORMS Conference on Business Analytics and Operations Research. This premier Analytics conference draws 800+ practitioners for three days of networking, professional development & intensive learning. For Sponsorship and Exhibitor details visit: http://meetings.informs.org/analytics2017

Contact: Olivia Schmitz INFORMS Exhibit & Sponsorship Sales Manager Olivia.Schmitz@informs.org 443.757.3539

CAESARS PALACE, LAS VEGAS APRIL 2-4, 2017 14 | ORMS Today

December 2016

ormstoday.informs.org

Is your solver provider

getting in the way of your success?

At Gurobi, we make it easier to achieve your optimization goals. Faster solutions with superior algorithms and powerful tuning Maximum productivity with flexible interfaces and modeling language support Great technical support from easy-to-reach optimization experts No surprises with our transparent pricing and flexible licensing

Gurobi Optimizer 7.0 State-of-the-Art Mathematical Programming Solver Experience state-of-the-art optimization algorithms backed by industry-leading customer support.

www.gurobi.com +1 713-871-9341 info@gurobi.com

INFORMS Initiatives

Student competition, logo, aCAP & more Student Analytical Scholar Case Study Competition INFORMS will once again offer the SAS/INFORMS Analytics Section Student Analytical Scholar Competition, a scholarship program that will send the winning recipient to the 2017 INFORMS Conference on Business Analytics and Operations Research. Supported by SAS and sponsored by the Analytics Society of INFORMS, the competitive program will recognize one outstanding student who would like to learn more about the practice of analytics at the conference in LasVegas on April 2-4, 2017.The scholarship covers the cost of attending the event and additional networking opportunities. The purpose of the competition is to practice the process of structuring and presenting a compelling proposal for analytical work. Applicants will be asked to produce a “Statement of Work” (SOW) for a case study based on a real-life project. Such documents are usually created early in a project, after some exploratory work, but may or may not fully define the problem. SAS will provide a case study. A discussion forum will be available Jan. 30-Feb. 10, 2017, for applicants, who are encouraged to ask questions and explore the problem definition in order to put together a viable, professional SOW.The winner will be selected based on the cohesion, proper use of assumptions and demonstrated technical and presentation skills in the SOW, as well as the practicality of the proposed approach to solve the business problem. The deadline to apply is Feb. 13, 2017. For more information, visit: informs.org/ SAS-AnalyticsScholarCompetition. INFORMS unveils new logo During the recent 2016 INFORMS Annual Meeting,Vice President of Marketing and Outreach Laura Albert McLay unveiled a new logo that will officially begin to be used with the launch of the new INFORMS website in the spring of 2017. 16 | ORMS Today

December 2016

McLay also previewed the new IOL (INFORMS Online), which is expected to go live prior to the 2017 Conference on Business Analytics and Operations Research in April. INFORMS launches aCAP The Certified Analytics Professional (CAP®) program has helped hundreds of organizations find experienced analytics professionals through a rigorous certification process. Now the program is helping organizations tap early career professionals with the launch of aCAP (Associate Certified Analytics Professional). The aCAP provides a trusted and independent verification of analytics knowledge of someone who is in the early stages of their career, allowing hiring managers to more easily identify the best possible analytics talent that will continue to grow and contribute to their organization. Likewise, aCAP provides emerging analytics professionals with a competitive advantage over their peers with the distinction of a certification, while affording them the opportunity to build the experience and soft skills necessary for a full CAP certification. Both the CAP and the aCAP programs are managed by INFORMS. For more information, visit https://www. certifiedanalytics.org/. Carefully chosen words boost fundraising donations A forthcoming study in the INFORMS journal Marketing Science, based on the psychology of sympathy, shows that small changes in the wording of a fundraising letter can increase donations by more than 300 percent. With more than a million registered public charities in the United States,

fundraising for good causes has become more difficult than ever.Annual events like Giving Tuesday increase overall giving, but also increase the competition for funds around those events. For their research, the authors of the study, K. Sudhir ofYale University, Subroto Roy of the University of New Haven, and Mathew Cherian of HelpAge, India, found that leveraging psychological theories on sympathy when drafting a fundraising letter can increase donations enormously. To test their hypotheses, the authors conducted a large-scale direct mail fundraising experiment on a cold list of about 200,000 potential donors across India and a warm list of more than 100,000 past donors on behalf of one of India’s most wellrespected charities that serve the elderly. The main findings were surprising. Donations changed dramatically based on key characteristics of the target of the donation and the appeal. On the cold list, donations went up by 110 percent if the target was a named individual versus an unnamed group, by 55 percent if the target belonged to the same religion as the donor versus a different religion, by 33 percent if the target fell into poverty versus being poor with an undescribed past, and by 66 percent if the annual donation was framed as monthly versus daily amounts. Combining all these tactics led to a 300 percent increase in donations. 2017 INFORMS Analytics Conference accepting submissions The deadline for submitting proposal presentations for the 2017 INFORMS Conference on Business Analytics & Operations Research is Jan. 15, 2017. The conference will be held April 2-4 in Las Vegas. Speakers selected through the approval process will receive a discount off regular registration rates. The Conference Selection Committee welcomes proposals in all areas and all topics within the business analytics and O.R. arena. Both presentation content and speaking expertise will be considered in selection, with priority given to real-world business topics and high-quality academic work geared to real situations. For more information, visit: http:// meetings2.infor ms.org/wordpress/ analytics2017/abstracts-submissions/. ORMS ormstoday.informs.org

Stop compromising. With FICOâ&#x20AC;&#x2122;s advanced analytics platform, you can deploy your existing math models with what-if analysis and visualization in a matter of weeks. Supported by 60 years of proven industry experience, you can provide trusted insights to increase productivity and improve profitability.

Roundtable Profile

Google’s headquarters in Mountain View, Calif. As one employee put it, Google is a “candy store” for O.R. practitioners (inset). Source: Google

Operations research at Google Advanced analytics permeates work at Google, making the multinational technology giant a ‘candy store for O.R. practitioners.’

By Brian Thomas Eck and Amber Richter 18 | ORMS Today

December 2016

The mission of Google, Inc. is “to organize the world’s information and make it universally accessible and useful.” This has spawned efforts as diverse as optical fiber to the home (gFiber), longevity research (Calico), smar t home automation (Nest), YouTube, glucose-detecting contact lenses (Verily), self-driving cars and many others. Such broad-reaching innovations are possible due to robust search and ads businesses. It is well known that web search was the foundation of Google. Recognition that a web page is important if it is pointed to by other important pages translates into math: PageRank’s importance scores are the stationary values of an enormous Markov chain [1]. With this start, it is not surprising that Google’s culture goes hand in hand with analytical literacy. ormstoday.informs.org

Working as an operations research (O.R.) practitioner surrounded by highly analytical colleagues brings, by contrast, new meaning to the term “isolated practitioner.” Wandering around Google, one sees whiteboards everywhere, filled with equations, graphs, pseudo code and probability distributions.Widespread respect for data to inform decisions is accompanied by healthy skepticism; numbers can also mislead. For example, peer reviews and presentations at all levels have the primary intent of improving the analysis. A presentation where the audience politely listens and applauds at the end represents a failure to engage. A successful presentation features frequent interruptions, challenges to assumptions and analysis, and lively debate with the audience.This holds true even in executive presentations, where the O.R. is vetted with deep experts in computer science and statistics, technical minds imbued with a broad grasp of the business. In contrast to organizations where advanced analytics methods remain shielded, Googlers pry open the Black Box and engage. For experienced hires from other companies, this can be initially disconcerting, but over time, they discover that they gain trust and impact by embracing this method of collaboration. Organizing information at scale often relies on software. Each day, Google’s systems crawl 20 billion web pages, stream hundreds of millions of hours of YouTube videos and activate 1.5 million Android devices. This scale requires a massive physical infrastructure: Google’s unparalleled worldwide cluster computing system.This infrastructure includes 13 data center campuses of staggering size.The Council Bluffs, Iowa, campus is the largest in the world, with multistoried data center buildings that have building pads over a third of a mile long. In addition, Google has presence in dozens of cities across more than 33 countries, with a global network of fiber optic cables connecting it all, to bring information and services quickly and reliably to its end users. Building and growing this infrastructure requires insights provided by hundreds of advanced analytics projects. Ongoing operations require efficient allocation of compute, storage and network resources across internal product areas and external cloud customers. Advanced analytics also make up the core function for many Google products, from improving users’ search results to finding an optimal driving route for Google Maps directions. Advanced analytics techniques go beyond what we traditionally define as O.R., and include methods from fields such as statistics, robotics, control systems, game theory, econometrics and risk analysis. For example, machine learning (ML) is used to improve search results, automate language translation, protect Gmail and Chrome users from spam and malware, and even improve data center energy efficiency (Google is the

All About the Roundtable The Roundtable consists of the institutional members of INFORMS with member company representatives typically the overall leader of O.R. activity. The Roundtable is composed of about 50 organizations that have demonstrated leadership in the application of O.R. and advanced analytics. The Roundtable culture is peer-to-peer, encouraging networking and sharing lessons learned among members. The Roundtable meets three times a year. Roundtable goals are to improve member organizations’ OR/MS practice, help Roundtable representatives grow professionally and help the OR/MS profession to thrive. Further information is available at http:// roundtable.informs.org. The Roundtable also has an advisory responsibility to INFORMS. According to its bylaws, “The Roundtable shall regularly share with INFORMS leadership and advise the INFORMS Board on its views, its suggested initiatives and its implementation plans on the important problems and opportunities facing operations research and the management sciences as a profession and on the ways in which INFORMS can deal proactively with those problems and opportunities.” The Roundtable meets with the INFORMS presidentelect each spring to discuss practice-related topics of interest to him or her, and with the entire INFORMS Board each fall to discuss topics of mutual concern. This series of articles aims to share with the INFORMS membership at large some information and insights into how O.R. is carried on in practice today.

largest corporate purchaser of renewable energy on the planet; Google’s data centers are among the most energy efficient in the world). Google has published hundreds of papers related to ML (see http://research. google.com) and has open-sourced many ML tools through TensorFlow (see www.tensorflow.org). A Vibrant Community of Quants O.R. practitioners are often interested in how companies organize their O.R. employees: in central teams, embedded within functional domains or some hybrid. Google uses a hybrid approach and augments it by providing clear direction on how individuals’ careers can advance, building and sustaining a community of quantitative analysts, and retaining that community’s identity. At Google, operations researchers need to be generalist problem solvers, and they typically work in roles such as data scientist (quantitative analyst), software engineer or research scientist. As such, they are held to the standards of their associated job ladder.These ladders describe expectations at each level throughout a contributor’s career. Through committee-based decision-making, the ladders provide consistency across interviewing, hiring, calibrating performance ratings and evaluating promotions.This discipline sets a uniformly high bar for hiring and promotion across the company, and the consistent expectations facilitate rotation among teams. Furthermore, because all professionals participate, they become deeply familiar with each other’s work. The data scientist, or quantitative analyst, ladder includes several hundred analysts, most with a statistics background, a significant minority with an O.R. background, and small groups from fields such as biostatistics, economics and computational engineering. The number of analysts supporting a domain can vary from a few to a few dozen. The ladders enable that degree of domain specialization December 2016

ORMS Today

| 19

Roundtable Profile while preserving consistently high standards for technical hiring and work and embedding the analyst in a broader technical community. In addition to job ladders, Google uses forums for professionals to share their work freely within the company, such as informal lunch series, tech talks, a data science blog and more formal global summits. By providing these community-building activities and job Technician inside a Google data center. Source: Google ladders, Google sustains community identity and career direction for its O.R. analysts while positioning O.R. ODS also does forecasting and capacity planning. practitioners in both centralized and embedded ODS produces a range forecast of the fleet, which conteams. sists of compute, storage and power capacity needs in The following section highlights two centralized the data centers.This is used in making many downteams: one with a functional focus on the technical stream decisions including acquisition of land and infrastructure domain and the other with a focus on utilities, new construction and network capacity augmethods and tools used across multiple application mentation. Moving from point forecasts to quantifidomains. cation of the variation implied by forecast error, and using this variation to set inventory buffers, necessarily Core O.R. Teams involves substantial organizational transformation as Operations Decision Support (ODS): This Mountain well as analysis. As in most companies, this integration View, Calif.-based team is comprised of operations of hard and soft skills is an essential ingredient in the research Ph.D.s who focus on Google’s technical intoolkit of an O.R. practitioner at Google. frastructure: optimizing the hardware supply chain, Beyond these examples, ODS applies advanced planning data center and wide-area network capacanalytics to optimize Google’s fleet. It uses mixed inteity, optimizing server deployments and lifecycles, and ger programming (MIP) models to plan server deployimproving the utilization of compute and storage ments across the fleet and to optimally add and reshape resources. Many of the projects are variants of wellcompute and storage capacity within each cluster of known trade-offs to optimize cost: the Newsvendor machines. ODS also uses simulation and machine problem, timing for technology refresh and determinlearning models to overcommit and schedule compute ing build frequency economic order quantities. ODS’s and storage capacity to improve utilization. focus on cost optimization led to its strong reputation for total cost of ownership management. Operations Research Team (O.R.): While the For example, Google positions network gear ODS team is organized around application domains, in multiple cities around the world in order to the O.R. team is organized around methods. This connect with peers (Internet Service Providers) Paris-based group develops and supports combinacloser to their end users. How many and which torial optimization software and applies it to largefacilities should be used, and which gear should scale, real-world problems across the company. This be placed where, require trade-offs between fasoftware engineering and research team originated cility and fiber costs to connect gear across sites. out of a challenge posed by Google Street View. The team uses simulation to cost-optimize straObtaining Street View imagery requires effitegic roadmaps for evolution of peering support ciently routing cars down streets around the world within Google’s network. to capture all the needed images. Solving this clasAnother example is deciding when to replace sic Chinese Postman Problem led to savings on an older server with one from the newest generalabor and car maintenance, reduced emissions and tion, which requires optimizing various costs. Analymore up-to-date imagery through shorter and thus ses such as this inform many thousands of decisions, more frequented routes.This problem motivated the some in the form of a policy, some as a simple calcufounding of the O.R. team as Google’s in-house velator, and some as a complex decision support tool, hicle routing team, and the team’s expertise quickly run either periodically or on demand. expanded from there. 20 | ORMS Today

December 2016

ormstoday.informs.org

The team develops its optimization software libraries to handle the speed, scalability and security that Google-scale projects demand. More than 150 teams at Google use these libraries, and most of them have been open-sourced as the or-tools suite, available on GitHub. These libraries include a gold-medal winning constraint solver, vehicle routing library, linear optimization solver, Boolean optimization solver, knapsack solver and libraries for solving flow and assignment problems (see https:// developers.google.com/optimization/). Although it grew out of StreetView, the O.R. team works on projects all across Google.The team has developed optimization algorithms to stabilize YouTube videos, direct navigation for the Loon (Internet balloon) fleet, and even assign people across Google to serve on promotion committees.The O.R. team has also worked with Terra Bella, Google’s subsidiary formerly known as Skybox Imaging. Terra Bella has satellites that orbit the earth in short cycles and capture high-resolution satellite imagery of places all around the world. Fixed orbit paths limit when locations are in view of each satellite, and data downlink opportunities are available only when the satellites are near fixed ground stations.The OR team developed a MIP approach to schedule the timing and location of satellite captures of target images and downlinks of satellite image data. Isolated Practitioners (Not) In addition to the large O.R.-focused teams, there are many individual and small groups of O.R. contributors all across Google linked together by the community mechanisms described above. Several O.R. practitioners work across Google Express, Google’s online delivery service providing fast delivery of products from popular retailers.They solve problems such as demand forecasting, capacity planning, scheduling and routing to help deliver products from retailers to customers. For example, some practitioners work on forecasting the number of orders by time of day and location to be able to schedule drivers and store operators via optimization algorithms that account for constraints such as the very short lead time of orders, staff preferences and consistency in individual staff schedules over time. An O.R. problem that arises often in Google infrastructure is dynamic, multi-dimensional bin packing and load balancing. One example is job scheduling in Google’s massively parallel computing environments. Here, the multidimensional items are jobs that need to be placed on machines (bins) subject to multiple hard and soft constraints, such as available CPU and RAM, job preferences, priorities and specialized hardware needs. The infrastructure-related Algorithms and Analytics teams work

with the relevant engineering teams to improve both online dynamic algorithms and offline MIPbased solutions for scheduling jobs, adding resources to data centers and answering related capacity planning questions. The Large-Scale Optimization research team, based in New York, works with the relevant engineering teams to improve the efficiency and robustness of Google’s computational infrastructure, such as the backend systems that serve search and Google’s external cloud offering. For example, the team applied balanced graph partitioning algorithms to cluster search terms according to how often they co-occur in search queries, then used this clustering to govern how queries are distributed among machines in the search backend. This change greatly increased the rate at which queries can be served via improved caching. A software engineering team in Network Architecture does capacity planning and risk analysis for Google’s wide area network of fiber optic cables. Their models seek to minimize cost while ensuring availability, speed and scalability, three key components of Google’s network. They use MIP models to determine the cheapest network that can route flows during a given set of fiber failure scenarios. A Monte Carlo simulation tests the resulting network against availability and latency service level requirements to determine additional failure scenarios to include in the MIP in the next iteration.

There are

always new problems to solve and

new impacts to deliver. The relevance

of O.R. and advanced analytics is

stronger than ever in this burgeoning

high-tech industry.

O.R. is Everywhere In summary, advanced analytics permeates work at Google. It might seem easier to describe where it hasn’t been applied. But this impression is quickly contradicted by nontraditional cases such as human resources identifying an optimal number of candidate interviews or a job posting for a food service analytics and insights manager.There are always new problems to solve and new impacts to deliver. The relevance of O.R. and advanced analytics is stronger than ever in this burgeoning high-tech industry. Working here is perhaps best summed up by a recent quote from a Google analyst:“Google is like a candy store for O.R. practitioners.” ORMS Brian Thomas Eck, Ph.D., and Amber Richter, Ph.D., are quantitative analysts on the Operations Decision Support team within Technical Infrastructure. Eck is also the Google representative to the INFORMS Roundtable. Disclaimer: The opinions expressed in this article are those of the authors and do not necessarily represent the views of Google.

REFERENCE 1. Langville, A.N. and Meyer, C.D., 2006, “Google’s PageRank and Beyond: The Science of Search Engine Rankings” (page 31), Princeton University Press, Princeton, N.J.

December 2016

ORMS Today

| 21

Utility companies operate thousands of miles of pipelines carrying flammable natural gas, creating a tradeoff of risk mitigation and an increase in rates. Image © Сергей Сергеев | 123rf.com

Rolling up

operational risk at PG&E

Emerging discipline provides an overall risk snapshot that allows diverse stakeholders to assess tradeoffs between safety, reliability and cost.

By Jordan Alen, Christine Cowsert Chapman, Melissa Kirmse, Farshad Miraftab and Sam Savage

22 | ORMS Today

December 2016

Utility companies operate thousands of miles of pipelines carrying flammable natural gas as well as power lines containing high-voltage electricity. The industry is highly regulated and must weigh risk mitigation investments against increases in utility rates.This places these firms in an environment where diverse stakeholders (regulators, rate payers, line repair crews, etc.) need to understand the tradeoffs between cost and various categories of risk, including financial, safety and reliability. It would benefit the industry to work toward a standardized consolidated risk statement that aggregates operational risk across the physical assets of the system in the manner that a consolidated financial statement aggregates the financial health of the firm. Risks Weren’t Additive – Until Now There has been no practical way to add risks together to consolidate risk until recently. Here’s why: consider an adverse event that has a 10 percent chance of injuring someone (or equivalently injures 1/10th of a person on average). ormstoday.informs.org

Now suppose the organization is exposed to 10 such risks. Because averages can be added, risk management procedures based on expected values estimate the total risk by multiplying 1/10th by 10 to get an average of one injury across the 10 events. Although this does give the correct average, risks are not characterized by averages but by extreme outcomes, such as 10 people being injured across the 10 events. Reducing the risk to an average is wrong because it ignores the degree of statistical dependence or independence between events.This is analogous to playing craps with a pair of “averFigure 1: Average dice (left) and real dice (right). age” dice, which can only display 3 1/2 dots (Figure 1). known as SIPmath, without add-ins or macros [6]. In fact, an average of 1/10 of an injury per event To leverage these developments, ProbabilityMancould represent extremely disparate outcomes: anyagement.org, a 501(c)(3) nonprofit, has created open, where between one chance in 10 of 10 injuries cross-platform standards and tools to help aggregate across the 10 events (if the results are completely risk calculations systemwide [7, 8]. So today, instead correlated) to one chance in 10 billion of 10 injuries of large monolithic risk simulations, SIPs of various (if they are completely independent). risks (generated on multiple platforms at multiple An alternative to rolling up risk with averages levels of the enterprise) may be added, multiplied is to use simulation. Traditionally, this has required or used in other calculations, and then rolled up to a single monolithic simulation that spans the entire higher levels. organization. These systems have been effectively Large firms have begun to develop internal employed in the insurance industry and financial communities of practice around probability engineering, but they involve significant investment. management, and the discipline was recently recognized Like all large programs, they run the risk of by Gartner Inc. as “transformational” [9].The time for collapsing under their own weight or of being too the consolidated risk statement has arrived. inflexible to maintain. No wonder so many risk A Hypothetical Risk Statement management procedures are ineffective [1, 2, 3]. H oweve r, g ove r n m e n t a g e n c i e s mu s t Figure 2 displays a fully “rolled up” Excel model of a regulate risk in areas as diverse as banking, food, conceptual consolidated risk statement (CRS) for a pharmaceuticals, transportation and utilities. This is multi-level organization with three categories of opnot just an academic exercise. When a regulatory erational risk. Financial risk accounts for the direct agency underestimates risk, it jeopardizes public monetary consequences of facility and equipment safety. When it overestimates risk, it imposes failure measured in millions of dollars over the upexcessive mitigation costs, penalizes the consumer coming year. Safety risk is measured in the number and stifles the economy [4]. of injuries over the same period, and reliability risk is measured in terms of minutes of lost service per A Modular Approach customer. The effects of various mitigations may be The emerging discipline of probability management tested using check boxes, which instantly run simcommunicates uncertainties as arrays of trials called ulations of 1,000 trials to display the residual risk SIPs, which may be added up across simulations post-mitigation. like numbers [5]. This means that enterprise-wide The model is based on a hypothetical SIP library risk simulations may now be broken down into of the joint probability distribution of risks across an Lego block-like modules created on a wide varientire enterprise. Because the output SIPs are simety of software platforms, and then rolled up. Reply named spreadsheet ranges, they may be used in cently, native Microsoft Excel has become powerful any Excel formula.Thus, it is easy to include a wide enough to perform calculations with SIPs, a process array of probabilistic outputs. At the fully rolled up December 2016

Reducing the risk to

an average is wrong because it

ignores the degree of

statistical dependence or independence

between events.

ORMS Today

| 23

Probability Management

Figure 2: Top level of a hypothetical consolidated risk statement.

level we have included only three, which contain the chances that pre-specified limits will be breached in each area of risk. Conditional formatting has been applied that turns the cells green when the chance is zero, and red when the chance is 50 percent. The authors suggest that you download this model at ProbabilityManagement.org, since vicarious simulation is not as good as the real thing. Figure 3 displays the fully â&#x20AC;&#x153;drilled downâ&#x20AC;? view of the same model exposing the multiple levels of the organization. Rolling up and drilling down are accomplished with the Group and Ungroup tools on the Data Ribbon in Excel. Ultimately, mitigation decisions will involve tradeoffs between the risk dimensions, which must accommodate the risk tolerances of diverse stakeholders. In this model we have incorporated a graphical method used by Doug Hubbard to display such tolerances [10]. Exceedance and tolerance chart. The graphs at the bottom of the CRS show an exceedance graph in blue and a risk tolerance level in red. The blue

Figure 3: Drilled down view of consolidated risk statement.

24 | ORMS Today

December 2016

curve is a direct output of the model and displays the chance that the risk will exceed the number shown on the X axis. The red curve is not an output, but an indication of the organizationâ&#x20AC;&#x2122;s risk tolerance. Where the blue curve is below the red curve, risk is within tolerance; where it is above the red curve, the risk is beyond tolerance. Coming up with such curves is at the crux of risk management. In fact, an important function of the CRS is to facilitate discussions among the stakeholders and to negotiate a mutually agreeable set of tolerance curves. For example, in this hypothetical model, replacing equipment reduces reliability risk, but due to the added use of repair crews, it increases their safety risk. To satisfy the interests of both customers and repair crews, this model rewards combining safety training with equipment repair. Risk Management is a Journey, Not a Destination Technologically we have reached a tipping point at which a consolidated risk statement can be created with everyday computers and software. But how does an organization start down this new path? Pacific Gas & Electric (PG&E) is a utility company that provides natural gas and electricity to roughly 16 million customers in northern and central California. For them, the journey toward consolidating risk began in September 2010, when a tragic gas pipeline explosion in San Bruno, Calif., became a catalyst for change and improved utility risk management practices. In 2011, a California Senate bill [11] mandated that PG&E, the operator of the pipeline, adopt a safety focus and risk-based decision-making framework. The California Public Utilities Commission (CPUC) required that pipeline operators subject to CPUC rate regulation develop a specific plan to identify and minimize hazards and systemic risk to protect the public and employees when proposing prog rams in their rate case applications. As a result, PG&E was directed to move away from characterizing risks as single values and begin using distributions of event outcomes. On the Electric side of the business, ormstoday.informs.org

an external consultant leveraged the recent improvements in Excel to develop simulations to improve the reliability of electric transmissions lines. Meanwhile, the Gas Operations team, under the guidance of another consultant, started developing its own models using the latest generation of free tools available at ProbabilityManagement.org.These tools help create SIPmath models, but all it takes is Excel to run them, so both electric and gas models are compatible from a probability management perspective. This article describes the experience with the Gas Operations team. In the past, operational risks have often been evaluated in isolation, for example, the risk of a transformer fire or the loss of containment of gas pipe. However, for cost-effective risk reA tragic gas pipeline explosion in San Bruno, Calif., became a catalyst for change and improved utility risk duction, it would be desirable to management practices. allocate a total budget to a portImage ÂŠ Eric Broder Van Dyke | 123rf.com folio of mitigation activities to minimize total risk across the various categories. areas. As more models were built, more questions The introduction of the third generation of free were raised, which led to more models, which led SIPmath Modeler Tools [12] in 2016 allowed Gas to more questions. In many cases, this resulted in Operations to employ rapid prototyping on a large models that had nothing to do with the goal of risk number of conceptual models. This led to quick quantification, but instead focused on improved dewins, and built credibility within the organization. cision-making, improved data visualization or meaPG&E managers exposed to this work, whether suring effectiveness of programs. they were new to the company or executives with Should the development of a consolidated risk decades of experience, could easily interact with the statement be done top down (estimating probability models and grasp the concepts being presented. distributions using subject matter experts and hisThese models often shed light on unrelated torical statistics) or bottom up (simulating failures

An asset level model This model (available for download at ProbabilityManagement.org) aggregates financial, safety and reliability risk across the assets making up a pipeline. Each asset has three SIPs that can be rolled up to view overall risk using the Excel data filter and subtotal formula. Using the filter function of Excel, the user can view the distribution of impacts under specified conditions. For example, one could select the total risk across all Fittings in the North region by adjusting the filters in cells D6 and H6. The model also displays percentiles along with a threshold of risk and associated likelihood of exceedance.

Figure 1: The model aggregates financial, safety and reliability risk across the assets making up a pipeline. December 2016

ORMS Today

| 25

Probability Management • “Interactive blueprints” are greatly simplified working models of the entire system.

Figure 4: Interactive blueprint of the end-to-end corrosion mitigation system.

The

Gas Operations team arrived at a

six-phase probabilistic framework to roll up

external corrosion risk across

PG&E’s gas transmission pipeline network.

26 | ORMS Today

through root cause analysis)? The answer is both. An organization must be comfortable with each approach, and the two are synergistic. Bottom up models can inform the probabilistic assumptions at the top level, while top down models can pinpoint areas in which additional detailed analysis is most beneficial. Following is an overview of a current initiative to optimize the mitigation of external corrosion on transmission pipes. A sidebar story describes how distributions of risks simulated on an asset level (that is, across pipes, valves, fittings, etc.) may be filtered to generate conditional distributions across portions of the system, for example, the sum of all risks across pipes in the North. The end-to-end process. Gas system assets are typically buried underground, and their conditions, operations and performances are highly uncertain. Many methods exist to quantify risk or create distributions of adverse outcomes for gas assets at the component or segment level. (e.g., a valve has a probability of sticking closed with a distribution of subsequent hours of service interruption, or a specific segment of pipe has a likelihood of failure with a distribution of potential safety consequences). The Gas Operations team arrived at a sixphase probabilistic framework to roll up external corrosion risk across PG&E’s gas transmission pipeline network, which after component testing is now under implementation. Instead of being developed from purely written plans, the effort is based on functioning prototype models. The four categories are: • “Paper airplanes” provide the team with a common understanding of some mathematical principles relating to risk. • “Balsa airplanes” look like real models in that they have user interfaces and data requirements suitable for managerial decisions. • “Eiffel Towers” are huge models whose purpose is to prove how big they can get before the limits of scalability cause them to collapse. December 2016

Figure 4 displays an interactive blueprint of the external corrosion system. It moves through a sixphase process that begins with a database of histor ical asset conditions and concludes with an optimized set of mitigation portfolios. In practice, each of the six phases would be a separate model, database or application, perhaps not even created on the same platform. Each would take input SIPs from the previous phase and pass output SIPs to the next. Phase 1 – initial conditions: This is a database describing the conditions of the assets as of the date of last inspection. The data can be as detailed as the records of thousands of small corrosion pits detected by sensitive electronic monitoring equipment run through the pipes. It also contains information on the locations and soil conditions of the pipes, as well as densities of the populations surrounding them. Phase 2 – time dependence: Probabilistic corrosion growth models estimate the current and future conditions of the assets. Phase 3 – loss of containment simulation: Adverse events such as leaks or ruptures are simulated based on the current conditions and external factors such as earthquakes. Phase 4 – conditional consequences: Distributions of consequences are simulated based on SIPs of adverse events generated in Phase 3. Phase 5 – mitigation strategies: The results of various mitigation strategies are simulated. Phase 6 – optimization: Optimization is performed graphically in the blueprint. In practice, stochastic optimization would be applied to the SIPs generated in Phase 5, but for this small model it is instructive to keep it interactive.The scatter plot displays cost vs. risk reduction for each combination of mitigations. The green dot displays the currently selected portfolio. An efficient frontier is observable in the southwest region of the graph. Conclusion In the past year, PG&E has developed probabilistic models in native Excel within both Electric and Gas Operations.The open SIPmath standard assures that these models may be used collaboratively to provide an overall risk snapshot that allows diverse stakeholders to assess tradeoffs between safety, reliability and cost. ormstoday.informs.org

As George Bernard Shaw said, “The single biggest problem in communication is the illusion that it has taken place.” This is particularly true in the area of enter pr ise r isk management. A consolidated risk statement would create a common language of risk that would make risks transparent to everyone involved. ORMS Jordan Alen is a technology coordinator at ProbabilityManagement.org and a risk consultant with experience in utility regulation and aggregated simulation models. Christine Cowsert Chapman is the senior director of Asset Knowledge & Integrity Management in Gas Operations at Pacific Gas and Electric Company (PG&E). She is responsible for developing the strategic direction of PG&E’s risk, asset and integrity management programs that are applied to all of PG&E’s natural gas assets. Melissa Kirmse is director of Operations at ProbabilityManagement.org. She has more than 20 years of experience in project coordination, administration, as well as technical writing and editing at tech companies such as Microsoft and TiVo. Farshad Miraftab is the senior risk analyst for Gas Operations Risk Management at PG&E. He is responsible for developing data-driven approaches to identify and mitigate operational risks as well as investment optimization solutions that prioritize Gas Operations portfolio based on risk, financial and other operational constraints.

REFERENCES 1. Sam Savage, 2009, “The Flaw of Averages,” John Wiley & Sons. 2. Douglas Hubbard, 2009, “The Failure of Risk Management: Why It’s Broken and How to Fix It,” John Wiley & Sons. 3. Philip Thomas, Reidar B. Bratvold and J. Eric Bickel, 2014, “The Risk of Using Risk Matrices,” Society of Petroleum Engineers, SPE Economics & Management, Vol. 6, Issue 2, April 2014. 4. Stephen Breuer, 1993, “Breaking the Vicious Cycle: Toward Effective Risk Regulation,” Harvard University Press. 5. Sam Savage, Stefan Scholtes and Daniel Zweidler, 2006, “Probability Management,” OR/MS Today, February 2006, Vol. 33, No. 1. 6. Sam L. Savage, 2012, “Distribution Processing and the Arithmetic of Uncertainty, Analytics Magazine, November/December 2012 (http://viewer.zmags.com/ publication/90ffcc6b#/90ffcc6b/29). 7. Melissa Kirmse and Sam Savage, 2014, “Probability Management 2.0, OR/MS Today, October 2014, Vol. 41 No. 5 (http://viewer.zmags.com/publication/ad9e976e#/ad9e976e/32). 8. Sam L. Savage, 2016, “Monte Carlo for the Masses,” Analytics Magazine, September/October 2016 (http://analytics-magazine.org/monte-carlo-for-the-masses/). 9. https://www.gartner.com/doc/3388917/hype-cycle-data-science10. Douglas Hubbard and Richard Seiersen, 2016, “How to Measure Anything in Cybersecurity Risk, p. 47, John Wiley & Sons. 11. California SB-705 Natural Gas: Service and Safety, October 2011. 12. http://probabilitymanagement.org/tools.html. Sam L. Savage, Ph.D., is the executive director of ProbabilityManagement.org, author of “The Flaw of Averages – Why We Underestimate Risk in the Face of Uncertainty,” and an adjunct professor at Stanford University. He is the inventor of the open SIP data structure that allows simulations to communicate with each other across platforms.

2017 ANALYTICS CAREER FAIR Find the Right Analytics Professional to Make Sense of Your Data. • Find the seasoned professionals you need – over 800 analytics professionals expected • Provide your recruitment materials in a casual setting • Arrange discreet on-site meetings in private booths • Enhance your visibility with an ad in Analytics or OR/MS Today

QUESTIONS?

CAREER CENTER

careers@informs.org or call (800) 4-INFORMs

http://meetings.informs.org/analytics2017 December 2016

ORMS Today

| 27

In a divided nation, the unprecedented 2016 presidential election was difficult to predict. Image © delcreations | 123rf.com

The election that

confounded everybody How the ‘13 Keys’ model outperformed its creator. By Douglas A. Samuelson

28 | ORMS Today

December 2016

What an election! Only two prominent forecasters got it right, and one of those only with some after-the-fact spin. Actually, three, if you count film director Michael Moore. Maybe. Perhaps a fourth, if we stretch a bit.We’ll get back to that. Professor Allan Lichtman’s “13 Keys” model [1] was spot-on once again, in one of the most hard-to-forecast presidential elections in living memory. Well, sort of. Actually, Lichtman got one key wrong and missed the popular vote result, which is what his model is supposed to forecast, but.... Oh, OK, it’s complicated. Patience, please. ormstoday.informs.org

As reported in Analytics magazine last July [2], at the time the Keys model looked too close to call. It uses 13 yes-no variables (see box) that reflect satisfaction with the incumbent party. It’s a statistical pattern recognition, a kernel discriminant function analysis [3]. As of July, three keys were still undetermined: incumbent-party contest (Key 2), third party (Key 4), and foreign policy success (Key 11). By mid-September, however, Lichtman felt sure enough to make his prediction. The persistence of the Bernie Sanders effort through the convention, garnering more than one-third of the delegate votes, “probably should have” tipped Key 2 against Clinton, but Sanders’ endorsement of Clinton before the convention left Lichtman doubtful about it. “I had six other keys,” Lichtman explained, “so I didn’t think much more about it.” Lichtman, a quantitative histor ian and a professor of history at Amer ican University in Washington, D.C., decided that the Iran nuclear deal, the Paris climate change accord and the shrinking territory held by ISIS were not enough to hold Key 11 – foreign policy or military success – in Clinton’s favor, as these achievements did not seem to have generated widespread popular acclaim. This meant that Gary Johnson’s third-party candidacy, consistently polling over 5 percent – actually, over 10 percent in September – was enough to turn the sixth key negative. (The other negative keys were Key 1, House seats; Key 3, incumbency; Key 7, policy change; and Key 12, incumbent party candidate’s charisma.) Therefore, Lichtman forecast a Trump win in September, when Clinton was leading comfortably in the polls. Success, right? Ah, but wait! The Keys model forecasts the popular vote, not the electoral vote. Clinton won the popular vote – narrowly, but she won, as much of her lead evaporated in the last two weeks before the election. But also during that time, Gary Johnson’s vote slipped below five percent – he wound up with about 3.7 percent. So the model was right, after all, but not in the way Lichtman interpreted it. However, as Lichtman is quick to point out, predictive models have to be correct ahead of time, not in retrospect. He called Key 4 against Clinton when Johnson was polling above 10 percent, consistent with how he has interpreted the third-party key in past years. “You’ll notice that this time I didn’t distinguish between the popular and electoral vote,” he added. “I just predicted that Trump

The 13 Keys to the Presidency 1. After the midterm election, the incumbent party holds more seats in the U.S. House of Representatives than it did after the preceding midterm election. 2. The incumbent-party nominee gets at least two-thirds of the vote on the first ballot at the nominating convention. 3. The incumbent-party candidate is the sitting president. 4. There is no third-party or independent candidacy that wins at least 5 percent of the vote. 5. The economy is not in recession during the campaign. 6. Real (constant-dollar) per capita economic growth during the term equals or exceeds mean growth for the preceding two terms. 7. The administration achieves a major policy change during the term, on the order of the New Deal, the first-term Reagan “revolution” or the Affordable Care Act. 8. There has been no major social unrest during the term, sufficient to cause deep concerns about the unraveling of society. 9. There is no broad recognition of a scandal that directly touches the president. 10. There has been no military or foreign policy failure during the term, substantial enough that it appears to undermine America’s national interests significantly or threaten its standing in the world. 11. There has been a military or foreign policy success during the term substantial enough to advance America’s national interests or improve its standing in the world. 12. The incumbent-party candidate is charismatic or is a national hero. 13. The challenger is not charismatic and is not a national hero. If six or more of these statements are false, the incumbent party loses. would win, because that’s what the keys indicated. In close elections, the Democrats always have an edge in the popular vote because of the huge margins they can run up in a few big states, notably California and New York. But the electoral vote doesn’t always follow the popular vote that closely any more. So far, nobody has commented on how that relationship has changed.” At least Lichtman deserves credit for consistently insisting the election would be close. More important, this election indicates more strongly than ever that general satisfaction or dissatisfaction with how things are going eventually overwhelms specific events in the campaign. It also marks, in Lichtman’s opinion, a major change in the election dynamics. “Truth, your record, demeaning large groups within the populace, even clear illegal acts don’t seem to matter,” he adds. “Herman Cain had to withdraw in 2012 because three women accused him of December 2016

Lichtman

deserves credit for consistently

insisting the election would

be close.

ORMS Today

| 29

Politics & Analytics

The

populace is increasingly divided into

“bubbles,” wherein most people

rely on news outlets that share a

strong point of view.

30 | ORMS Today

sexual harassment, which he denied. Trump had more than a dozen women accusing him of sexual harassment, plus a video of him bragging about how he’d done it, and he just kept right on going. All the precedents are out the window. If a candidate like Trump can win, anyone can.” Why the Polls Were Wrong On the other hand, one thing this election clearly does is cast serious doubt on the polls. “Polls are snapshots,” Lichtman says dismissively. “They’re poor predictors.” But this year was worse than most others in recent memory, with changes of 4 percent or 5 percent between the actual vote and polls taken less than a week before the election.The most likely explanation for the disparity between the polls and the actual result, according to a number of pollsters and political scientists including Lichtman, is that the pollsters’ estimates of respondents’ likelihood of voting were seriously wrong. John Zogby, a senior partner at John Zogby Strategies and one of the most accurate pollsters for the past 30 years, saw the problem coming. Three weeks before the election [4], despite Hillary Clinton’s double-digit leads in most polls at the time, he declared, “I can’t tell you who’s really going to win. Tell me who will vote, and I’ll tell you who will win. If we get around 132 million votes, as we did in 2008 and 2012, Hillary wins. If we get 121 million, as we did in 2004, Trump wins.” We actually got around 126 million, giving Clinton about a 2 million vote lead. Shortly after Election Day, with 122 million votes counted, her lead was closer to 200,000. Several million non-voters, particularly Democratic-leaning people in the north central states, helped tip those states to Trump. So Zogby’s assessment looks quite accurate. Zogby also asserted, a week before the election, that the memo from FBI Director James Comey was a non-factor. The week of Oct. 23-28, an ABC poll showed Clinton leading Trump by 14 points, but on Monday, Oct. 31, it turned into a 1-point lead for Clinton, and on Tuesday, a 1-point lead for Trump just a week before voters went to the polls (except, of course, those who had already voted early). “Very simply [Clinton has] been in a downward spiral for the last nine days, so this was not an overnight thing or a Comey thing. This has been a slow and steady decline by Hillary,” Zogby said Nov. 1, a week before the election, to Steve Malzberg on “America Talks Live.” Another person who called it right, surprisingly, was film director Michael Moore, who said in July December 2016

that anger and frustration among white male skilled workers, mostly union members, in the north central industrial states could tip those states – Wisconsin, Michigan, Ohio and Pennsylvania – away from Hillary Clinton. He was far from alone – according to some unconfirmed reports, the state Democratic chairs and Bill Clinton were trying to convey the same message to Hillary and urging her to push in those states with a strong economic message. She had never appeared in Wisconsin during the general election campaign, and she had skipped the traditional Labor Day Democratic campaign rally in Cadillac Square in Detroit. Lichtman is skeptical that it made a difference in the outcome: “Strategic campaign decisions, like whether to campaign hard in Wisconsin, are only right or wrong in retrospect. If she’d carried Wisconsin, Michigan and Arizona you’d be calling her a genius.” Again, those campaign decisions occur within a context of general satisfaction or dissatisfaction with how things are going in the country, and those metrics of general satisfaction or dissatisfaction, according to the Keys model, drive the result. The New Tribalism There are a couple of other factors that have apparently escaped notice so far. Zogby, who now focuses more on business targeting, sees a marked tendency toward tribalization in America: that is, Americans identify more and more with virtual tribes of like-minded people, and less and less with region, party or other traditional organizational structures [5]. Interestingly, he did not apply his tribalism analysis to the election, but he does see the relevance. One of the reasons he got the turnout effect more accurately than most pollsters is that he used, among other sampling frames, weekly Walmart shoppers. Certainly many commentators have noted that the populace currently is increasingly divided into “bubbles,” wherein most people rely on news outlets that share a strong point of view, and they associate mostly with other people who share that point of view. This tendency goes along with a rapidly declining number of competitive congressional districts, as reapportionments tend to create mostly safe districts for one party or the other.This, in turn, means that representatives run more toward the “true believer” base voters, to make sure they win their primaries, and have little incentive to move toward the center as one would in a swing district. This also means that strong partisan views by voters get reinforced rather than broadened or moderated by their representatives. ormstoday.informs.org

This change in the structure of electorate is closely related to change in communications. Many social media sites have algorithms that direct to the user more content similar to what they have been viewing, reinforcing insularity and strong partisan views. Possibly most polls, still relying heavily on telephones and email, missed the shift in some parts of the populace to a reliance on social media, especially Twitter.There may also have been a decline in some voters’ willingness to respond to polls at all. Trump, a master salesman, used social media more effectively than Clinton, both for messaging and for analytics. Previous articles in Analytics and OR/MS Today [6, 7] recounted President Obama’s success in using social media. Although Hillary Clinton and the Democrats learned much from Obama and others, the question of how well they used social media and related analytics will no doubt feed some lively discussions, both among political staffs and scholars, for the next few years. Marshall McLuhan Called It Looking again at the increasing insularity of large blocs of voters, the reinforcing effects various media and the decreasing importance of policy issues, it seems that this election also provided dramatic confirmation for one more analyst: Marshall McLuhan. As early as 1964, in his penetrating analysis of the 1960 election, he noted the ascendance of style over content because of the new visual medium (TV) and foretold the stillaccelerating changes in politics [8]. People who listened to the Kennedy-Nixon debates on the radio thought Nixon won. People who watched on TV thought Kennedy won. McLuhan explained that Kennedy was visually appealing and harder to classify – and that advantage overwhelmed any disadvantage on specific issues. This phenomenon helps to explain the voters’ indifference to issues and misstatements. McLuhan said that TV is “an extension of the sense of touch,” steering attention away from “linear” content. McLuhan asserted, “Anyone whose appearance strongly declares his role and status in life is wrong for TV.” Applying this insight to the 2016 campaign, we can see that Hillary Clinton’s appearance all but screamed, “corporate lawyer turned politician.” As McLuhan said of Nixon, the ease of classifying Hillary left the viewer vaguely uneasy about her without being able to articulate why. Donald Trump, at first glance seemingly too “hot” and harsh for TV, did keep the viewer guessing about who he was and what he might say next – and that promotes visual engagement,

the key to success. He was not a reality TV star by coincidence, nor was that experience irrelevant to the campaign. So once again the American people awoke the morning after the election wondering who the president-elect really was – a predictable consequence of McLuhan’s theories [9]. McLuhan also predicted that the emergence of new media would change the structure of society (“The Medium Is the Message”) and produce a form of tribalism on a global scale – the “Global Village,” as he called it. Now we see it. The next question is what to do with it. Perhaps a new coalition-building approach, building on the emerging tribal structure, will prove successful. Or perhaps personal visual appeal will prevail over content, and our future leaders will be TV personalities first, with exper ience counting less, if at all. Or there may be a reaction against the current situation, producing a renewed emphasis on proven political experience. The 13 Keys model still implies that successful governance determines who wins the presidency, notwithstanding all the other phenomena. Who knows? It seems highly likely that those political leaders, opinion leaders and analysts who first figure it out will be the ones who will succeed in the future. ORMS

The

13 Keys model still

implies that

successful governance determines

who wins the presidency, notwithstanding all the

other phenomena.

Doug Samuelson (samuelsondoug@yahoo.com), a frequent contributor to OR/MS Today, is president of InfoLogix, Inc., a consulting company in Annandale, Va. Samuelson worked as a paid campaign staffer in a U.S. Senate campaign in Nevada in 1970, as a county coordinator in a gubernatorial campaign and targeting analyst for a local campaign in California in 1974, and as a Federal Civil Service policy analyst from 1975 to 1982. More recently, Samuelson worked for the Election Science Institute on the analysis of discrepancies between the exit polls and the actual vote in Ohio in 2004 and on the trials of new voting machines in 2006.

REFERENCES 1. Allan J. Lichtman, 1988, “The 13 Keys to the Presidency”; 2016 ed., “Predicting the Next President: The Keys to the White House,” Rowman and Littlefield, Lanham, Md., 2. Douglas A. Samuelson, 2016, “Politics & Analytics: Who Holds the Keys to the White House?” Analytics, July-August. 3. Douglas A. Samuelson, 1996, “Unlocking the Door to the White House,” OR/MS Today, October. 4. John Zogby, 2016, “Election 2016: How Did We Get to This?” webcast, Center for Strategic and International Studies, www.csis.org, Oct. 19. 5. John Zogby, 2016, “We Are Many, We Are One: Neo-Tribes and Tribal Analytics in 21st Century America,” John Zogby, New Hartford, N.Y. 6. Douglas A. Samuelson, 2013, “Analytics: Key to Obama’s Victory,” OR/MS Today, February. 7. Douglas A. Samuelson, 2014, “Analytics Penetrates Deeper into Politics,” OR/MS Today, October. 8. Marshall McLuhan,1964; revised, critical edition, 2003, “Understanding Media: The Extensions of Man,” Gingko Press, Corte Madera, Calif. 9. Douglas A. Samuelson, 2008, “Marshall McLuhan’s Parable,” OR/MS Today, December.

December 2016

ORMS Today

| 31

Are orange cars the least likely lemons?

Sound

data science Avoiding the most pernicious prediction pitfall.

By Eric Siegel

32 | ORMS Today

December 2016

Data science and predictive analytics’ explosive popularity promises meteoric value, but a common misapplication readily backfires. The number crunching only delivers if a fundamental – yet often omitted – fail-safe is applied. Prediction is booming. Data scientists have the “sexiest job of the 21st century” (as Thomas Davenport and U.S. Chief Data Scientist D.J. Patil declared in 2012). Fueled by the data tsunami, we’ve entered a golden age of predictive discoveries. A frenzy of analysis churns out a bonanza of colorful, valuable and sometimes surprising insights [1]: • People who “like” curly fries on Facebook are more intelligent. • Typing with proper capitalization indicates creditworthiness. • Users of the Chrome and Firefox browsers make better employees. • Men who skip breakfast are at greater risk for coronary heart disease. • Credit card holders who go to the dentist are better credit risks. • High-crime neighborhoods demand more Uber rides. ormstoday.informs.org

Look like fun? Before you dive in, be warned: This spree of data exploration must be tamed with strict quality control. It’s easy to get it wrong, crash and burn – or at least end up with egg on your face. In 2012, a Seattle Times article led with an eye-catching predictive discovery: “An orange used car is least likely to be a lemon” [2]. This insight came from a predictive analytics competition to detect which used cars are bad buys (lemons).While insights also emerged pertaining to other car attributes – such as make, model, year, trim level and size – the apparent advantage of being orange caught the most attention. Responding to quizzical expressions, data wonks offered creative explanations, such as the idea that owners who select an unusual car color tend to have more of a “connection” to and take better care of their vehicle. Examined alone, the “orange lemon” discovery appeared sound from a mathematical perspective. The specific result is shown in Figure 1. According to Figure 1, orange cars turn out to be lemons one-third less often than average. Put another way, if you buy a car that’s not orange, you increase your risk by 50 percent. Well-established statistics appeared to back up this “colorful” discovery. A formal assessment indicated it was statistically significant, meaning that the chances were slim this pattern would have appeared only by random chance. It seemed safe to assume the finding was sound. To be more specific, a standard mathematical test indicated there was less than a 1 percent chance this trend would show up in the data if orange cars weren’t actually more reliable. But something had gone terribly wrong. The “orange car” insight later proved inconclusive. The statistical test had been applied in a flawed manner; the press had ran with the finding prematurely. As data gets bigger, so does a potential pitfall in the application of common, established statistical methods. The Little Gotcha of Big Data “The trouble with the world is that the stupid are cocksure and the intelligent are full of doubt.” – Bertrand Russell

Big data brings big potential – but also big danger. With more data, a unique pitfall often dupes even the brightest of data scientists. This hidden hazard can undermine the process that evaluates for statistical significance, the gold standard of scientific soundness. And what a hazard it is! A bogus discovery can spell disaster.You may buy an orange

Figure 1: Are orange cars really less likely to turn into lemons?

car – or undergo an ineffective medical procedure – for no good reason. As the aphorisms tell us, bad information is worse than no information at all; misplaced confidence is seldom found again. This peril seems paradoxical. If data’s so valuable, why should we suffer from obtaining more and more of it? Statistics has long advised that having more examples is better. A longer list of cases provides the means to more scrupulously assess a trend. Can you imagine what the downside of more data might be? As you’ll see in a moment, it’s a thought-provoking, dramatic plot twist. The fate of science – and sleeping well at night – depends on deterring the danger. The very notion of empirical discovery is at stake.To leverage the extraordinary opportunity of today’s data explosion, we need a surefire way to determine whether an observed trend is real, rather than a random artifact of the data. How can we reaffirm science’s trustworthy reputation? Statistics approaches this challenge in a very particular way. It tells us the chances the observed trend could randomly appear even if the effect were not real.That is, it answers this question [3]: Question that statistics can answer: If orange cars were actually no more reliable than used cars in general, what would be the probability that this strong of a trend – depicting orange cars as more reliable – would show in data anyway, just by random chance?

This peril

seems paradoxical. If data’s

so valuable, why should we suffer

from obtaining more and more of it?

With any discovery in data, there’s always some possibility we’ve been “Fooled by Randomness,” as Nassim Taleb titled his compelling book. The book reveals the dangerous tendency people have to subscribe to unfounded explanations for their December 2016

ORMS Today

| 33

Predictive Analytics The experimenters’

mistake was to

not account for running many

small risks, which had

added up to

one big one.

own successes and failures, rather than correctly attributing many happenings to sheer randomness. The scientific antidote to this failing is probability, which Taleb affectionately dubs “a branch of applied skepticism.” Statistics is the resource we rely on to gauge probability. It answers the orange car question above by calculating the probability that what’s been observed in data would occur randomly if orange cars actually held no advantage. The calculation takes data size into account – in this case, there were 72,983 used cars varying across 15 colors, of which 415 were orange [4]. The calculated answer: under 0.68 percent. Looks like a safe bet. Common practice considers this risk acceptably remote, low enough to at least tentatively believe the data. But don’t buy an orange car just yet – or write about the finding in a newspaper for that matter. What Went Wrong: Accumulating Risk “In China when you’re one in a million, there are 1,300 people just like you.” - Bill Gates

So if there had only been a 1 percent long shot that we’d be misled by randomness, what went wrong? The exper imenters’ mistake was to not account for running many small risks, which had added up to one big one. In addition to checking whether being orange is predictive of car reliability, they also checked each of the other 14 colors, as well as the make, model, year, trim level, type of transmission, size and more. For each of these factors, they repeatedly ran the risk of being fooled by randomness. Probability is relative, affected entirely by context. With additional background information, a seemingly unlikely event turns out to be not so special after all. Imagine your friend calls to tell you, “I won the jackpot at hundred-to-one odds!”You might get a little excited. “Wow!” Now imagine your friend adds, “By the way, I’m only talking about one of 70 times that I spun the jackpot wheel.” The occurrence that had at first seemed special suddenly has a new context, positioned alongside a number of less remarkable episodes. Instead of exclaiming “wow,” you might instead do some arithmetic.The probability of losing a spin is 99 percent. If you spin twice, the chances of losing both is 99 percent x 99 percent, which is about 98 percent. Although you’ll probably lose both spins, why stop at two? The more times you spin, the lower the chances of never winning once. To figure out the probability of losing 70 times in 34 | ORMS Today

December 2016

a row, multiply 99 percent times itself 70 times, aka 0.99 raised to the power of 70. That comes to just under 0.5. Let your friend know that nothing special happened – the odds of winning at least once were about 50/50. Special cases aren’t so special after all. By the same sort of reasoning, we might be skeptical about the merits of the famed and fortuned. Do the most successful elite hold talents as elevated as their singular status? As Taleb put it in “Fooled by Randomness,” “I am not saying that Warren Buffett is not skilled; only that a large population of random investors will almost necessarily produce someone with his track records just by luck.” Play enough and you’ll eventually win. Likewise, press your luck repeatedly and you’ll eventually lose. Imagine your same well-intentioned friend calls to tell you, “I discovered that orange cars are more reliable, and the stats say there’s only a 1 percent chance this phenomenon would appear in the data if it weren’t true.”You might get a little impressed. “Interesting discovery!” Now imagine your friend adds, “By the way, I’m only talking about one among dozens of car factors – my computer program systematically went through and checked each one.” Both of your friend’s stories enthusiastically led with a “remarkable” event – a jackpot win or a predictive discovery. But the numerous other less remarkable attempts – that often go unmentioned – are just as pertinent to each story’s conclusion. Wake up and smell the probability. Imagine we test 70 characteristics of cars that in reality are not predictive of lemons. But each test suffers, say, a 1 percent risk the data will falsely show a predictive effect just by random chance. The accumulated risk piles up. As with the jackpot wheel, there’s a 50/50 chance the unlikely event will eventually take place – that you will stumble upon a random perturbation that, considered in isolation, is compelling enough to mislead. The Potential and Danger of Automating Science: Vast Search “The most exciting phrase to hear in science, the one that heralds new discoveries, is not ‘Eureka!’ but rather ‘Hmm… that’s funny’…” – Isaac Asimov

A tremendous potential inspires us to face this peril: Predictive modeling automates scientific discovery. Although it may seem like an obvious thing to do in this computer age, trying out each predictor variable is a dramatic departure from the classic scientific method of developing a single hypothesis and then testing it. Your computer ormstoday.informs.org

ASSOCIATE CERTIFIED ANALYTICS PROFESSIONAL Analyze How aCAP Can Help Launch Your Analytics Career

The Associate Certified Analytics Professional (aCAP) is an entry-level analytics professional who is educated in the analytics process but may not have practice experience. This prestigious program begins a career pathway leading to the CAP designation.

www.certiďŹ edanalytics.org

Predictive Analytics essentially acts as hundreds or even thousands of scientists by conducting a broad, exploratory analysis, automatically evaluating an entire batch of predictors. This aggressive hunt for any novel source of predictive information leaves no stone unturned.The process is key to uncovering valuable, unforeseen insights. Automating this search for valuable predictors empowers science, lessening its dependence on ever-elusive serendipity. Instead of waiting to inadvertently stumble upon revelations or racking our brains for hypotheses, we rely less on luck and hunches by systematically testing many factors. But as exciting a proposition as it is, this automation of data exploration builds up the risk of eventually being fooled – at one time or another – by randomness. This inflation of risk comes as a consequence of assessing many characteristics of used cars, for example.The power of automatically testing a batch of predictors may serve us well, but it also exposes us to the very real risk of bogus discoveries. Let’s call this issue vast search – the term that industry leader John Elder coined for this form of automated exploration and its associated peril. Repeatedly identified anew across industries and fields of science, this issue is also called p-hacking or the multiple comparisons trap [5]. Elder warns,

“The problem is so widespread that it is the chief reason for a crisis in experimental science, where most journal results have been discovered to resist replication; that is, to be wrong!” Statistics darling Nate Silver jumped straight to the issue of vast search when asked generally about the topic of big data on Freakonomics Radio: “You have so many lottery tickets when you can run an analysis on a [large data set] that you’re going to have some one-in-a-million coincidences just by chance alone.” Bigger data isn’t the problem; more specifically, it’s wider data. When prepared for predictive analytics, data grows in two dimensions – it’s a table (see Table 1). As you accrue more examples of cars, people or whatever you’re predicting, the table grows longer (more rows, aka training cases). That’s always a good thing. The more training cases to analyze, the more statistically sound [6]. Expanding in the other dimension, each row widens (more columns) as more factors – aka predictor variables – are accrued. A certain factor such as car color may only amount to a single column in the data, but since we look at each possible color individually, it has the virtual effect of adding 15 columns to the width, one per color. Overall, the sample data in Table 1 is not nearly as

One row per predictor variable Wider means more predictors – vaster search

Dodge

Neon

2004

Compact

Silver

Mitsubishi

Galant

2004

Medium

White

Mercury

Sable

2004

Medium

White

BAD

Ford

Focus

2005

Compact

Silver

Kia

Spectra

2004

Medium

Black

Dodge

Caravan

2005

Van

Red

BAD

Ford

Explorer

2002

Medium

Blue

BAD

Chrystler

Pacifica

2004

Crossover

Silver

Pontiac

Vibe

2004

Medium

Orange

...

One row per training case: Longer means more cases – better

Table 1: Data for predicting bad buys among used cars. The complete data is both wider and longer.

36 | ORMS Today

December 2016

ormstoday.informs.org

wide as data often gets, but even in this case the vast search effect is at play. With wider and wider data, we can only tap the potential if we can avoid the booby trap set by vast search. A Failsafe for Sound Results “There are three kinds of lies: lies, damned lies and statistics.” - Benjamin Disraeli

“If you torture the data long enough, it will confess.” -Ronald Coase

To understand what sort of failsafe mechanism we need, let’s revisit the misleading “orange lemons” discovery (Figure 1). The 12.3-vs.-8.2 result is calculated from four numbers: There were 72,983 cars, of which 8,976 were lemons.There were 415 orange cars, of which 34 were lemons. The standard method – the one that misled researchers as well as the press – evaluates for statistical significance based only on those four numbers. When fed these as input, the test provides a positive result, calculating there was only a 0.68 percent chance we would witness that extreme of a difference among orange cars if they were in actuality no more prone to be lemons than cars of other colors. But these four numbers alone do not tell the whole story – the context of the discovery also matters. How vast was the search for such discoveries? How many other factors were also checked for a correlation with whether a car is a lemon? In other words, if a data scientist hands you these four numbers as “proof ” of a discovery, you should ask what it took to find it. Inquire, “How many other things did you also try that came up dry?” With the breadth of search taken into account, the “orange lemon” discovery collapses. Confidence diminishes, and it shows as inconclusive. Even if we assume the other 14 colors were the only other factors examined, statistical methods estimate a much less impressive 7.2 percent probability of stumbling by chance alone upon a bogus finding that appears this compelling [7]. Although 7.2 percent is lower odds than a coin toss, it’s no long shot; by common standards, this is not a publishable result. Moreover, 7.2 percent is an optimistic estimate. We can assume the risk was even higher than that (i.e., worse) since other factors such as car make, model and year were also available, rendering the search even wider and the opportunities to be duped even more plentiful. Inconclusive results must not be overstated. It may still be true that orange cars are less likely

The

to be lemons, but the likelihood this would have appeared in the data by chance alone is too high to put a lot of faith in it. There’s not enough evidence to rigorously support the hypothesis. It is, at least for now, relegated to “a fascinating possibility,” only provisionally distinct from any untested theories one might think up. Want conclusive results? Then get longer data, i.e., more rows of examples. Adequately rigorous fail-safes that account for the breadth of search set a higher bar. They serve as a more scrupulous filter to eliminate inconclusive findings before they get applied or published. To compensate for this strictness and increase the opportunity to nonetheless attain conclusive results, the best recourse is elongating the list of cases. If the search is vast – that is, if the data is wide – then findings will need to be more compelling in order to pass through the filter. To that end, if there are ample examples with which to confirm findings – in other words, if the data makes up for its width by also being longer – then legitimate findings will have the empirical support they need to be validated. The potential of data will prevail so long as there are enough training examples to correctly discern which predictive discoveries are authentic. In this big data tsunami, you’ve got to either sharpen your surfing skills or get out of the water. ORMS

potential of data will prevail

so long as there are

enough training examples to

correctly discern which predictive

discoveries are

authentic.

Eric Siegel, Ph.D., is the founder of the Predictive Analytics World conference series (cross-sector events), executive editor of The Predictive Analytics Times and a former computer science professor at Columbia University. This article was adapted and reprinted with permission of the publisher, Wiley, from the book, “Predictive Analytics: The Power to Predict Who Will Click, Buy, Lie, or Die,” revised and updated edition, by Eric Siegel (Wiley, January 2016).

NOTES & REFERENCES 1. For more details on these findings, see the section on “Bizarre and Surprising Insights” within the Notes for the book, “Predictive Analytics,” available as a PDF online at www. PredictiveNotes.com. For further reading on this article’s overall topic, look in the section, “Further Reading on Vast Search” within the same document. 2. This discovery was also featured by The Huffington Post, The New York Times, National Public Radio, The Wall Street Journal and The New York Times bestseller, “Big Data: A Revolution That Will Transform How We Live, Work and Think.” 3. The notion that orange cars have no advantage is called the null hypothesis. The probability the observed effect would occur in data if the null hypothesis were true is called the p-value. If the p-value is low enough – e.g., below 1 percent or 5 percent – then a researcher will typically reject the null hypothesis as too unlikely, and view this as support for the discovery, which is thereby considered statistically significant. 4. The applicable statistical method is a one-sided equality of proportions hypothesis test, which calculated the p-value as under 0.0068. 5. More synonyms: multiple hypothesis testing, researcher degrees of freedom and cherry-picking findings. 6. This only holds true under the assumption you have a representative sample, e.g., an unbiased, random selection of cases. 7. This probability was estimated with a method called target shuffling. For details, see “Are Orange Cars Really not Lemons?” by John Elder and Ben Bullard (elderresearch.com/orange-car).

December 2016

ORMS Today

| 37

Q&A

Moving forward with membership help

INFORMS President-Elect Brian Denton

Incoming INFORMS President Brian Denton discusses goals, aspirations and member-inspired ideas and initiatives for 2017.

By Peter Horner

38 | ORMS Today

December 2016

Brian Denton recently recalled his introduction to INFORMS. At the time, he was a grad student, pursuing a Ph.D. in management science at McMaster University in Canada. He was in Montreal to present a paper at an INFORMS Annual Meeting, his first. A native of Burlington, Ontario, Canada, Denton remembers being ver y ner vous; he imagined a thousand people would at tend his session. Turns out about 980 fewer people showed up to hear his talk than he expected, but he left the conference full of excitement about his decision to pursue an OR/MS career and certain he had found his professional home in INFORMS. Fast forward two decades. On Jan. 1, 2017, Denton will become the president of INFORMS after serving the past year as president-elect. Along the way, he served INFORMS in many other roles, including two terms as Board secretary, as chair of the INFORMS Health Applications Section, as program chair for the INFORMS Annual Meeting and as chair of the Franz Edelman Award Committee. He also picked up his share of accolades from INFORMS, including the INFORMS Service Section Prize and the INFORMS Daniel H. Wagner Prize. Denton, who already held an undergrad degree in chemistry and physics (1994) from McMaster University and a master’s in physics (1996) from York University (Canada), completed his Ph.D. in 2001, and went to work as a senior engineer at IBM, followed by a twoyear stint as a senior associate consultant with the Mayo Clinic. He returned to academia as an assistant professor in the Department of Industrial & Systems Engineering at North Carolina State University in 2007. In 2012, he joined the faculty at the University of Michigan, where today he is a professor in the Department of Industrial and Operations Engineering. We interviewed Denton by phone on Nov. 8 and again a week later at the 2016 INFORMS Annual Meeting in Nashville. The conversations literally ranged from “A” to “Z,” from the state of the analytics profession and INFORMS, to his goals for his year as president, to his research interests in optimization and medical decision-making, to his fascination and fun with fly fishing and, of course, zombies. Following are excerpts: ormstoday.informs.org

What is the “state of INFORMS” from your viewpoint in terms of its major activities and its financial position? The state of INFORMS is excellent. We are in really good shape. We have a large reserve fund that is making it possible to initiate new activities. We have more than 12,000 members for the first time, and things are going great with our portfolio of member services. We have a large group of exciting meetings, our journals are in great shape, and we have all kinds of new member services we are creating along the way. Looking more broadly, what is the state of the greater analytics profession? I think public enthusiasm about analytics is creating all kinds of opportunity for our membership. Whether our members consider themselves operations researchers or management scientists or data scientists or something else, analytics is creating a lot of new opportunities. I think much of it is driven by the fact that there’s increasingly easy access to data, and that’s creating more and more opportunities to use some of the approaches we’ve been developing for a long time, and it’s becoming much more popular with the general public. It’s also creating opportunities to do new stuff. I think our members are benefiting in terms of having the opportunity to work on new problems and use real data to help make better decisions, to hire new faculty, and to find new business opportunities because of greater public awareness of our field. In your position statement, you said your first priority will be to focus on future strategic initiatives. What new initiatives are in the pipeline? Are you reaching out to the membership for ideas? Absolutely. One of the things I’m very interested in is helping members make some of their good ideas come to light. INFORMS recently developed a proposal process that allows any member to propose new ideas. It’s not a complicated process, and it can provide the opportunity to get funding to do something new. We want to get as many of these new proposals as we can so we can select the best ones to create some new opportunities. I’ve been spending time thinking about how to enhance this strategic proposal process and how to get the word out to as many people as possible. Can you talk about specific new initiatives? Sure, I’ll give you some recent examples I’m excited about. One is the development of ethical guidelines for our membership. I appointed Dave Hunt to a committee to lead this effort. Dave put

together an amazing committee of members, including former presidents and editors-in-chief of our journals, and developed some guidelines that are aspirational so our members have some kind of idea of what they should aspire to if they work in our profession. Another new initiative I’m very excited about is a diversity and inclusion initiative. Last January, I appointed an ad hoc committee with Michael Johnson as the chair. Michael recruited a great group of people to start developing some ideas around how we can lower the barriers to participation for our members, and to be more cognizant of our membership and the diversity of our membership. A proposal to make this a long-term sustainable effort was approved at the Board meeting in Nashville. There’s a lot more in the pipeline, but I won’t ruin all of the surprises.

One of the things I’m very

interested in is

helping members make some of their

good ideas come

I don’t suppose you grew up dreaming about being an operations researcher or management scientist. What led you from a bachelor’s degree in chemistry and a master’s degree in physics to a Ph.D. in management science, which ultimately became the foundation of your career? Who does? I think we all discover operations research along the way. I guess everybody has their own story of how they discovered operations research. In my case, I was interested in finding something where I could use what I had learned in a more practical way, something with more opportunities for employment, and that’s when I discovered operations research. I found it’s a great field to be in; it’s just a fantastic way to use some of the skills that people develop in a lot of different disciplines such as math or engineering or basic sciences. There’s all kinds of important problems in the area of operations research that our members can impact.

to light.

Presumably you had some mentors who guided you and influenced your O.R. journey. There are too many mentors to mention them all. I’ve really benefited from having many great mentors over my entire career. I will mention a few, though. One was my Ph.D. advisor, Diwakar Gupta at McMaster University. He really helped me to understand how to develop research ideas and how to do research. After my transition from grad school to my job at IBM, many people helped me, especially my IBM colleague John Milne, who showed me the ropes about how things work in a research and development environment. If I had to mention December 2016

ORMS Today

| 39

Q&A

two more people, they would be the first faculty mentors I had at North Carolina State University, Steve Roberts and Jim Wilson, both senior faculty members. Without their guidance I don’t think I would have been successful as an academic. What’s the best advice you’ve both received and could give a young person contemplating an O.R. career? Great question. For a young person, I can think of two very good pieces of advice. One is, operations research is such a broad area with so many opportunities, make sure you do something that you really care about and that you’re very excited about. Another piece of advice is not to listen to pessimists or negative people. Don’t let people tell you that your ideas are not any good. I guess the best advice I heard from one of my mentors was don’t avoid things because you’re worried about failing; failing is a natural outcome of learning. That was something that stuck with me early in my career.

[Academia] is actually very

As far back as your grad student days in the 1990s, you picked INFORMS as your professional home? What drew you to the Institute? I never even thought that it was a choice. At the time, my grad school peers and faculty were all talking about INFORMS, and when I went to IBM for my first job, I was encouraged to stay connected to INFORMS. The thing that I was most excited about at that time was the conferences. At the annual meeting, there were 50 or 60 tracks of talks all going on at once. It was like being a kid in a candy store – just an endless number of things to learn. Now, much later in my career, INFORMS is still helping me develop my professional network of collaborators and helping me meet people of similar interests to my own.

entrepreneurial;

there’s an enormous amount of freedom to

follow your research interests.

Do you remember your first INFORMS Annual Meeting? It was a meeting in Montreal, 1997 I believe, and I went there with a group of other grad students. I was giving a presentation, and I was both terrified and excited. I imagined there would be a thousand people in the audience. It turned out there were only about 20 people in my session, but it was still a great experience. I remember that distinctly. I was surprised at how comfortable I felt at the conference and how welcoming the people were. It changed from something that was maybe initially a scary experience to one where I really had a great time, and I wanted to go back to another INFORMS meeting as soon as I could. 40 | ORMS Today

December 2016

Following your Ph.D., you worked at IBM and the Mayo Clinic before joining the academic ranks. What did you take away from that real-world practice experience? B a s i c a l l y, I s t i l l t h i n k o f my s e l f a s a practitioner, and I still have a lot of the friends at those places. A lot of the research that I do here at the University of Michigan finds its way into practice. I guess that’s the thing I took away from IBM and the Mayo Clinic: do research from the perspective of someone who has done it as a practitioner. I took away tons of ways to think about problems and a general understanding of how the world works when it comes to trying to have an impact. Would you recommend that sort of realworld/academic combination experience? Does one help the other, and vice versa? Students ask me that all the time. I would say the industry experience has helped me a lot. But for somebody who wants to be an academic, if that’s their major career goal, they should approach going into industry first with caution. There’s not one kind of academic position, and there’s not one kind of practice position. Some industry research positions will allow you to develop a CV that would be attractive to academics, and some would not.Will you be able to present your work at conferences? Is publishing one of the requirements for the job? These are questions I would ask. Why did you decide to re-enter the academia world after several years in the practice world? Well, I discovered that academic positions are actually a lot different than what I imagined, which is pretty strange because my father is a professor. I always thought I knew what it was like to be a professor. You go and sit in an office and read books, teach, and once in a while students drop by and talk to you. But while I was at Mayo, and working with people in engineering schools, I discovered that my perception of what an academic’s life is like is quite different than reality. It’s actually very entrepreneurial; there’s an enormous amount of freedom to follow your research interests. There are many different ways to be a professor. You can be very theoretical or very applied or somewhere in between. I had the opportunity to learn that from some of the people I worked as external collaborators with while I was at the Mayo Clinic. I discovered I liked working with grad students a lot. I had a number of them come in as summer interns.Those are the things that changed my mind about academic positions, and then I was ormstoday.informs.org

very lucky to have an opportunity to go to North Carolina State University. So the combination of my interests and having the opportunity was really what made me make that transition. I really enjoyed my two years at the Mayo Clinic, but I found out an academic position really suited me and I could keep working with people at the Mayo Clinic, collaborators that I am working with to this day. Now I have grad students I’m sending to Mayo Clinic instead of advising them at Mayo Clinic. Your research is focused on optimization under uncertainty with applications to healthcare delivery and medical decisionmaking. You’ve also chaired the INFORMS Healthcare Section. What about the healthcare sector attracted you? Initially, I was interested in problems that were related to operations management and logistics such as planning and scheduling. That’s what I did during my Ph.D. thesis and during my time at IBM. Healthcare was always sort of an interesting, mysterious area to me; there didn’t seem to be as much operations research work in that sector compared to manufacturing, for example. So I learned something about healthcare logistics problems and found a position about it that was advertised in OR/MS Today. A big ad for Mayo Clinic positions. I sent off my CV. I didn’t expect them to contact me, but it turns out most people didn’t have a whole lot of experience in healthcare operations research dur ing those days. They talked to me about scheduling as related to surgery, and I admitted I had no experience in surgery scheduling. They were receptive to what I had learned in my grad studies and in working at IBM, they were convinced it was transferable, and they made me an offer. After I got there, I discovered that I was really interested in problems in the area of medicine, like how to screen for cancer and how to optimize treatment decisions for patients with the goal of trying to prevent heart disease or stroke. That would have never happened if I didn’t go to Mayo Clinic and get immersed in the topic area. Optimization in medicine is what I focus all my time on now. The healthcare industry is a target-rich environment for optimization, but it has also been historically reluctant to optimization. How do you get the buy-in? That’s a great question. I’ll try not to go on for hours. There are many different kinds of problems in the healthcare industry. There are very important

business problems, operation management problems and strategic capacity investment problems, and optimization can play a really important role in all of them. One of the challenges is not so much explaining the concept of optimization. The big challenge that I see is the politics. Maybe a better way to say it is there are many stakeholders, and they naturally have different criteria in mind for what makes for a great healthcare system. Surgeons, primary care physicians, nurses, administrators – they all have overlap on what they think is important, but they don’t always agree on what the most important thing is. So one of the big challenges is getting buy-in on a solution. If you think about optimization problems, we tend to think about an objective function in the simplest kind of context, but there’s obviously a lot of criteria, so they are very complicated, multiple-criteria optimization problems that require a lot of talking to people in order to understand what motivates their preferences. That’s on the operation side. On the medical side, there’s more agreement on what the right thing to do is. People have the patient’s best interest in mind, so I find it’s not hard to explain ideas about optimization. It’s consistent with how people think about things. The hard part is convincing them and yourself that the model is well validated and trustworthy. What’s critically important there is developing approaches that are based on real data and recognizing some of the challenges that exist, including missing data, various sources of biases, etc. Those are critically important to getting buy-in and influence decision-making.

It’s not

hard to explain ideas about

optimization. The hard part

is convincing them and yourself that

the model is validated and

trustworthy.

INFORMS has organized several conferences focused on healthcare, including the upcoming 2017 meeting in Rotterdam. Tell us about your involvement with the conference and the Healthcare Society of INFORMS. A huge number of INFORMS members are involved in healthcare in some way. I’ve lost track of the exact number of Healthcare Society members, but it’s in the high hundreds. I was chair of it during a period of transition, when we went from a section to a society, with below 500 members to well over 500. It was an exciting time. When I was chair of the Healthcare Section, tons of people were always saying, “We should have a healthcare conference.” So I put together a business case for the first healthcare conference. I mentioned to [then Director of Meetings] Terry Cryan that I thought there would be enough interest in healthcare to have a focused conference, and she said that fit pretty well with what they had December 2016

ORMS Today

| 41

Q&A

been talking about at the Board level. I wrote up a short business case, gave it to Terry and she took it to the Board and got people behind it. That led to the first conference in Montreal, and now it seems to be enduring. I’ll for sure be going to the conference in Rotterdam. The healthcare conference wasn’t really my idea; I’m just the one who conveyed it. There are a lot of members out there with great ideas, but maybe the Board doesn’t always hear them. That’s why I’m so excited about strategic initiatives and opening up the proposal process.

It’s a lot

easier to implement ideas that you

as a member have than you

might think. There’s

funding, there’s

people to help.

INFORMS is held in high esteem by the academic community because of its prestigious journals and conferences. How would you describe INFORMS’ standing or relevance outside the academic world? In the healthcare industry, for example, or the manufacturing industry? That’s a good question, and it’s something I spend a lot of time thinking about. We actually have an enormous amount going on all over the place. In medicine, for example, there are thousands and thousands of articles on using stochastic models and optimization. They aren’t appearing in INFORMS or operations research journals necessarily. The vast majority of these articles are appearing in medical journals. This is just one example. Operations research is permeating industries in ways that we are not necessarily aware of as a membership or as an institution. One thing that I think is really important, that [INFORMS President] Ed Kaplan and I and the Board have been talking about, is developing a means to monitor and discover the new ways that operations research is being used beyond our traditional journals and conferences. The goal is to shine a light on the many things that our members and our field is having an impact on, usually behind the scenes. We have a great opportunity to advertise the great work our members are doing all over the world. Let’s talk about volunteering. Why do you and many other INFORMS members devote so much time and energy to the Institute and the profession? Personally, I like most of all the opportunity to meet new people. By working with people on a volunteer project, whether it’s organizing a session at a conference or organizing a cluster or organizing a program for the annual meeting or whatever, you get to meet people and perhaps develop a lasting relationship. That’s something I really enjoy –

42 | ORMS Today

December 2016

meeting and working with interesting new people. By doing that, I can learn more about my research, more things that help my teaching in the classroom, but also more things I’m just generally interested to learn about out of curiosity. All of this is a fun for me, but it’s also helped my career in a lot of ways. I’ve developed a network of people that I can communicate with who might, for example, be able to help one of my grad students find a new position. Sometimes I get advice on projects I’m working on or collaborate with the people I meet through INFORMS. These are all direct benefits from volunteering. You’ve served a couple of terms on the INFORMS Board as secretary and now as president-elect. I would guess that volunteering has also bolstered your leadership skills. I’ve had amazing volunteer experiences, starting with organizing a session at a conference.You sort of work your way up a ladder, doing more things, developing your leadership skills and sometimes making mistakes.You learn from those mistakes. We have really helpful people in our membership who are happy to provide mentoring. Of course, being on the Board is an opportunity to learn leadership skills. During my time on the Board, those have included Rina Schneur, Terry Harrison, Anne Robinson, Steve Robinson, Robin Keller and now Ed Kaplan. It’s been an enormously valuable experience for me to watch and learn from them, and now I’m really excited about being president starting Jan. 1. Did you seek the presidency, or, like many of your predecessors, did you respond to a call of duty from the Nominating Committee? It was the second one: Steve Robinson from the Nominating Committee called me one afternoon and asked if I would consider running for president-elect. I was humbled that the committee would ask. It wasn’t a plan of mine, but when he asked, I didn’t have to think about it too long before I told him that I would be happy to run. Having now served many years on the Board, what would you want the membership to know about INFORMS that they probably don’t? One thing that comes to mind is this: It’s a lot easier to implement ideas that you as a member have than you might think.There’s funding, there’s people to help. Go back to the healthcare conference we talked about earlier. That was a member idea. You ormstoday.informs.org

can have more influence on the society than you think. That’s one thing I’d tell members and want them to know. So what do you like to do for fun when you’re not teaching, researching or volunteering your time for INFORMS? I like spending time with my wife, family and friends, obviously. One thing that people might not know about me is I really like fly fishing. There’s some great fly fishing in Canada; I go up there quite a bit. Michigan is also good. Lately, I’ve gone to the Bahamas to fish. I’ve got fishing holes scattered all around. Maybe after my year as president I might find time to learn how to tie my own flies. Another thing people probably don’t know about me is I’m a big fan of zombies. I follow the “Walking Dead” pretty closely; I’ve probably watched every episode twice. I even work zombies into some of my classes. I create optimization problems around zombies. What’s cool about zombies? Everything! In my class, if I give a problem on manufactur ing or eng ineer ing in some

My goal

gener ic context, and then I g ive the same problem in the context of zombies, students are all over it. They get more excited. I don’t know what it is. Complete the following sentence: “I would consider my year as president of INFORMS a success if … I don’t mess it up [laugh]. Honestly, if we can implement some of the ideas I talked about in my position statement – the diversity inclusion initiative, developing a strong repository of data for research and educational purposes, developing more exciting member services. If we can implement some of those things in a way that they are going to be enduring and have a positive impact on our membership, then I’d be really happy. Being president isn’t like running a 100meter dash; it’s more like a relay race. I’ll pick up the baton from Ed Kaplan and hand it to Nick Hall at the end of the year. My goal is to hold up my end and leave things in great shape for the future. ORMS

is to

hold up my end and

leave things in

great shape for the

future.

Peter Horner (peter.horner@mail.informs.org) is the editor of OR/MS Today and Analytics magazines.

SAVE THE DATE

HEALTHCARE 20 7

OPTIMIZING OPERATIONS & OUTCOMES July 26–28, 2017 Rotterdam, Netherlands

HTTP://meetings.informs.org/healthcare2017 December 2016

ORMS Today

| 43

Learn how analytics and O.R. can maximize the value of your data to drive better business decisions.

http://meetings2.informs.org/analytics2016

http://meetings.informs.org/analytics2017

ormstoday.informs.org

news 2017 Business Analytics & O.R. Conference set for Las Vegas

Inside News

46 47 47 48 49 49 50 51 51 51 52 52 53 53 54 56 56

INFORMS Fellows President’s Award Impact Prize Kimball Medal Expository Writing Award Undergraduate O.R. Prize von Neumann Prize Teaching OR/MS Prize Dissertation Award Nicholson Student Paper Doing Good O.R. Prize Volunteer Service Award People Tribute: Marius Solomon Photos from Nashville Snapshot Survey Meetings

INFORMS elections: Call for nominees

Las Vegas will host the 2017 INFORMS Conference on Business Analytics & O.R. on April 2-4. Image © Somchai Jongmeesuk | 123rf.com

The 2017 INFORMS Conference on Business Analytics and Operations Research will be held in Las Vegas on April 2-4 at Caesars Palace. Caesars is one of the most prestigious casino hotels in the world, as well as one of Las Vegas’ largest and bestknown landmarks. Analytics 2017 will bring together nearly a thousand leading analytics professionals and industry experts to share ideas, network and learn through real-life examples of data-based analytical decisions. Long-formatted talks offer an outlet to hear the complete story of successful analytical projects from inception through implementation. The conference also offers substantial networking opportunities, making it the analytics event of the year for anyone who works in the analytics, operations research or management science fields. IBM Solution Executive Maher Lahmar chairs the conference. Hand-picked topics and speakers: The conference has seen huge growth and success year after year due in part to the conference program committees, whose members develop the topic tracks, select

speakers and organize the presentations that comprise the heart of the conference. This year’s program committee includes analysts and managers from companies such as Accenture, BNSF Railway, Chevron, Deloitte, Gartner, Google, Innovative Decisions, Intel, InterContinental Hotels Group, Kroger, Lockheed Martin, Mayo Clinic, The MITRE Corp., SAS, Schneider and Walt Disney Company, as well as leading universities and government agencies. The conference committee has designated nine topical tracks for the 2017 invited speaker program: Analytics Leadership and Soft Skills, Analytics on Unstructured Data, Decision and Risk Analysis, Emerging Analytics, Entertainment and Gaming, Internet of Things, Marketing Analytics, Revenue Management and Pricing, and Supply Chain Applications. The program will be rounded out by six tracks of handpicked, member-contributed talks, software tutorials from vendor sponsors and poster presentations. Technology workshops: Analytics 2017 officially starts with the welcome 2017 Conference, continued on p. 46

This is a call for nominations for the INFORMS elections to be held in 2017 to elect members of the 2018 Board of Directors. Below is a brief review of the nomination and election process. More detail is provided in INFORMS Bylaw 3 and P&P 4.2 and 4.3. Each year, a call for nominations of candidates for next year’s INFORMS Board occurs. For the 2017 elections, the Board positions for 2018 to be filled include: • President-elect • Secretary • Vice president-International Activities • Vice president-Practice • Vice president-Membership/ Professional Recognition • Vice president-Marketing, Communications & Outreach The position of vice president-chapters/fora is also to be filled, but that position will be filled through election by the Subdivisions Council. INFORMS vice presidents are typically elected for a two-year term. The treasurer and secretary positions are also two-year terms, and they are staggered so that in each year not more than one of the two positions has a newly elected person starting a term of office. Because these offices involve a significant learning experience, first-term incumbents may be nominated without opposition for a second term for reasons of continuity and stability. This is the reason why one or several positions on the INFORMS ballot may have only a single candidate. Call for Nominations, continued on p. 56

December 2016

ORMS Today

| 45

n ews INFORMS welcomes 12 Fellows INFORMS honored 12 new Fellows for their “outstanding lifetime achievement in operations research and the management sciences” at a special luncheon during the 2016 INFORMS Annual Meeting in Nashville, Tenn. INFORMS Fellows have “demonstrated exceptional accomplishments and made significant contributions to the advancement of operations research and management science over a period of time.” The award, which brings together the very best operations researchers and analytics experts throughout the world, recognizes outstanding achievement in five areas: education, management, practice, research and service. The 2016 Fellow honorees include: Stephen P. Boyd, Stanford University, for exceptional teaching and broad dissemination of convex optimization and outstanding research leading to innovative formulations and algorithms for problems across a wide array of disciplines. Kevin Glazebrook, Lancaster University, for fundamental contributions to the theory and practice of operations research in the area of applied probability and for pioneering initiatives in doctoral education. Peter J. Haas, IBM Research Division, for sustained and fundamental contributions to discrete-event simulation and interactive sampling-based analytics for massive data sets, as well as for significant service to the simulation community. Jeff Linderoth, University of Wisconsin-Madison, for fundamental contributions to research in computational mathematical programming, specifically for mixed-integer programming, global and stochastic optimization and grid computing, and for leadership in all aspects of computing within operations research. Sanjay Mehrotra, Northwestern University, for contributions to continuous, discrete and stochastic optimization methodology and their impact on O.R. technology implementation and application. George J. Miller, Altarum Institute, for contributions to O.R. practice and models in military, environmental, emergency preparedness and healthcare areas with successful implementations and broad highlevel impact to humankind and society. 46 | ORMS Today

December 2016

Georgia Perakis, MIT, for exceptional research, spanning theory to practice with important contributions to variational inequalities, the price of anarchy, dynamic pricing and data analytics, and for her dedicated mentorship of a future generation of O.R. scholars. Mauricio G. C. Resende, Amazon.com, Inc., INFORMS President Ed Kaplan (far left) and Committee Chair for major contributions to Jim Dyer (far right) welcome the Class of 2016 Fellows during a special luncheon held in conjunction with the 2016 INFORMS the development and appliAnnual Meeting in Nashville, Tenn. cation of metaheuristics for optimization problems, including the invention of the widely used greedy Paolo Toth, University of Bologna, randomized adaptive search procedure. Italy, for his outstanding methodological Ariela Sofer, George Mason Uni- contribution to the field of combinatorial versity, for her contributions to nonlinear optimization and for the application of optimization and especially its application his knowledge to the solution of railway to medical diagnosis and treatment, to O.R. planning problems. education and for her leadership in and serPascal Van Hentenryck, University vice to INFORMS and to the profession. of Michigan, for his scientific contributions Tamás Terlaky, Lehigh University, for to constraint programming, discrete fundamental contributions to the theory, al- optimization, global optimization, local gorithms, computational methodology and search and stochastic optimization, as well applications of optimization, as well as his as his many contributions to the practice of exemplary mentorship and distinguished operations research and the education of service to the INFORMS community. future practitioners. ORMS

2016 Conference on Business Analytics and O.R. 2017 Conference, continued from p. 45

reception on Sunday evening, April 2, but you can get an early start with the technology workshops that are offered all day April 2. The no-cost workshops offer in-depth training in software solutions. You can sign up when you register for the conference. Po s t e r p r e s e n t e r s o f f e r e d discounted registration rate: Interested students and professionals may submit a poster proposal and receive a discounted registration rate if selected to present at the meeting. The poster format is great for works-in-progress on which the speaker is looking for feedback or successful projects that may not be extensive enough for an extended talk. The deadline for all poster presentations to be submitted is Feb.

20, 2017. If your poster is selected to be presented, you can take advantage of a discounted registration rate of $1,070. Save with early conference rates: Early rates of $1,175 for INFORMS members and $1,470 for nonmembers are available until Feb. 6, 2017. Organizations can take advantage of the $1,070 per person team discount rate when they send three or more attendees to the conference. A $1,070 newcomer rate is also offered. This special rate applies to any INFORMS member who is attending the conference for the first time. All meals for two days are included in all registration fees. For information regarding conference registration or submitting a presentation, visit meetings.informs.org/analytics2017. ORMS ormstoday.informs.org

ormstoday.informs.org

NPS Professor Brown receives INFORMS President’s Award Gerald G. Brown, Distinguished Professor of Operations Research at the Naval Postgraduate School, was named the 2016 recipient of the INFORMS President’s Award for his “his work to improve societal welfare through the identification of new problems, development of appropriate models, and implementation of operations research methods to some of the most pressing military and security issues of our time.” Brown has been a member of the NPS faculty since receiving his Ph.D. from UCLA in 1974. The award was presented by INFORMS President Ed Kaplan at a special awards session held at the INFORMS Annual Meeting in Nashville, Te n n . T h e I N F O R M S Pr e s i d e n t ’ s Award recognizes and encourages important contributions to the welfare of society by operations researchers at the local, national and global level. The award committee includes the current INFORMS president and the two most recent past presidents. The award citation read in part: Over his 40-plus year career, Professor Brown has established himself as the world’s leading expert in military operations research. More to the point of this award, however, are Professor Brown’s tireless efforts to improve security, both in the United States and abroad. Professor Brown’s and his NPS colleagues’ research contributions to methods for defending critical infrastructure in addition to other problems in military and homeland security have been widely recognized. For these as well as basic research accomplishments in optimization theory, Professor Brown was elected to the National Academy of Engineering in 2008. Perhaps less known to INFORMS members is Professor Brown’s role in military problems such as developing a route planning tool for military aircraft operations in Afghanistan and Iraq, a life-saving intervention recognized with the Distinguished Civilian Service Medal awarded by the Secretary of the Navy. Other military applications

include developing software for planning Tomahawk missile strikes and capital planning models for defense procurements. The President’s Award is given to recognize work that advances the welfare of society. Prof. Brown’s design of new tools and programs have improved both the efficiency and effectiveness of military and homeland security operations. What better advances the welfare of society than successful operations research that keeps us safe and secure? ORMS

INFORMS President Ed Kaplan (left)) and President’s Award recipient Gerald Brown.

Eight share Impact Prize for contributions to revenue management The 2016 INFORMS Impact Prize was awarded to Peter Belobaba, E. Andrew Boyd, Tom Cook, Guillermo Gallego, Robert Phillips, Barry C. Smith, Kalyan Talluri and Garrett van Ryzin for their pivotal role in the creation and widespread adoption of revenue management. The Michael Fry (far left) and INFORMS President Ed Kaplan prize, sponsored by Princeton (far right) congratulate Impact Prize recipients Garrett Consultants, was presented at van Ryzin, Barry C. Smith, Andrew Boyd, Robert Phillips the INFORMS Annual Meeting and Guillermo Gallego. in Nashville, Tenn., by committee chair Michael Fry. retailing, entertainment, consumer finance Awarded every other year, the and manufacturing. As the Internet and the INFORMS Impact Prize is intended to information economy has grown, revenue recognize contributions that have had a broad management has become even more critical impact on the field of operations research. to many businesses including Amazon, Airbnb, Google, Microsoft and Uber. The 2016 citation read in part: From its earliest days, the field of revenue Revenue management is the theory and management has been characterized by an practice of dynamically managing the price active interchange between industry and and availability of a portfolio of products or academia. Revenue management has proven services to maximize profitability. Revenue to be a deep and enduring topic for research. management was famously first developed There are two specialty journals devoted to to enable airlines to manage the prices and the field of revenue management, as well as availability of their seat inventory following a Pricing and Revenue Management Section of deregulation in 1978. During the 1980s it INFORMS and an annual INFORMS Revenue spread from the airlines to other capacity- Management and Pricing Conference. constrained service industries such as hotels, Furthermore, academic research has resulted rental cars and cruise lines. From the 1990s in approaches and algorithms that are routinely through today it has become a mainstream used to improve revenue management business practice across a wide range of systems used in industry that deliver additional industries including not only airlines and revenues to companies while enabling access hotels, but also freight transportation, media, of scarce resources to customers. ORMS December 2016

ORMS Today

| 47

n ews Camm, Lougee receive Kimball Medals for distinguished service Jeff Camm, associate dean of business analytics at Wake Forest University and formerly a faculty member for 31 years at the Lindner College of Business at the University of Cincinnati, and Robin Lougee, a research scientist at the IBM T.J. Watson Research Center and currently the global research industry lead for Consumer Products and Agriculture of the IBM Research Division, received the George E. Kimball Medal for distinguished service to INFORMS and the O.R. profession. The medals were presented at the 2016 INFORMS Annual Meeting in Nashville, Tenn. Committee member Rina Schneur presented the medals on behalf of committee chair Don Kleinmuntz. The citations read in part: Jeffrey D. Camm: Dr. Camm’s area of research is optimization applied to problems in operations management and marketing. His research has been published in Science, Management Science, Operations Research and Interfaces, among other journals. His work in supply chain optimization with Procter & Gamble was a 1996 Edelman Award Finalist and is credited with helping P&G save more than $250 million annually in its North American supply chain. In 1998, his joint work on nature reserve site selection for efficient conservation was published in the journal Science and appeared in a brief to President Clinton. Camm has also been the recipient of a number of teaching awards, including the 2006 INFORMS Prize for the Teaching of OR/MS practice. He is co-author of seven texts on business statistics, management science and business analytics. His teaching, heavily influenced by his extensive consulting experience, has always been focused on creating truly impactful O.R. practitioners. Camm has a long history of service to INFORMS. Early in his career, he was active in the Cincinnati/Dayton Chapter of INFORMS, including serving as secretary, vice president and president of the chapter. He served as functional editor of Operations Management for Interfaces from 1999-2004 and as editor in chief of Interfaces from 200548 | ORMS Today

December 2016

2010. He served as associate editor of INFORMS Transactions on Education from 1999-2013. An avid supporter of the Edelman Award, he has served as an award judge twice, as an Edelman coach, as a member of the Edelman Gala Committee and as the Edelman Gala Master of Ceremonies. He has served on the INFORMS Prize Committee, Distinguished service (l-r): Committee member Rina including one year as chair. Schneur, medal recipients Jeff Camm and Robin Lougee and INFORMS President Ed Kaplan. Camm has been particularly active in service to INFORMS meetings, having served as cluster chair, Annual to the Board of the COIN-OR Foundation, Meeting program chair and as a member of Inc. The initiative has grown from its initial the INFORMS Meetings Committee for 10 offerings of four software projects to more years. He currently serves on the INFORMS than 50 projects spanning much of computaCommittee on Master’s Programs in Analytics. tional operations research. In 2014, she was named a co-recipient of the INFORMS ImRobin Lougee: Dr. Lougee’s research pact Prize in recognition of her contributions has been in the areas of mixed-integer linear to COIN-OR. Her fellow recipients honored programming theory and its application to the pivotal nature of her contributions by askbusiness and industry. She has published ing her to accept the award on behalf of them journal articles, book chapters and conference and the greater COIN-OR community. proceedings in the areas of mathematical Lougee has been engaged in service to programming, discrete optimization, INFORMS across a broad range of activimanufacturing production planning, ties, but with a particular focus on building hydroelectric management, open-source INFORMS’ communities and meetings. software, and object-oriented applications She served on the INFORMS Board as VP frameworks in journals including Operations of Meetings for two years, following four Research, Operations Research Letters and the years as a member of the Meetings ComIBM Journal for Research and Development. mittee, during a period of record-setting She has served as an associate editor for attendance growth. She also served as proSurveys in ORMS and Operations Research, and gram chair for the INFORMS International as a guest editor of IBM Journal of R&D special Meeting in Puerto Rico in 2007. issues on smarter commerce. She drove the creation of the INFORMS One of Lougee’s many significant pro- Professional Colloquium for students, an fessional achievements is as co-creator of the initiative that is now an integral component Computational Infrastructure for Operations of the INFORMS Conference on Business Research (COIN-OR), an initiative to spur Analytics and Operations Research, serving the development of open-source software as its inaugural chair in 2005 and 2006. Her to accelerate the adoption and evolution of leadership in INFORMS communities includes computational operations research. In 2000, serving for eight years as an officer of the she gave the presentation that launched the INFORMS Computing Society, including as COIN-OR initiative as a three-year experi- chair, during which time she established an ment by IBM Research. As program manager, endowment for the Student Prize, and for four she led its growth to an independent non- years as an officer of the Women in ORMS profit in 2004. She was subsequently elected Forum, including as president. ORMS ormstoday.informs.org

ormstoday.informs.org

Glasserman presented with Saul Gass Expository Writing Award Paul Glasserman of Columbia University was named the 2016 recipient of the Saul Gass Expository Writing Prize. Committee Chair Shane Henderson announced the award at the INFORMS Annual Meeting in Nashville, Tenn. Named in honor of the late Saul Gass, an O.R. pioneer and an extraordinary and prolific writer, the prize recognizes an operations researcher/management scientist whose publications demonstrate a consistently high standard of expository writing. The citation read in part: Professor Glasserman has heavily influenced the operations research and operations management community with his research contributions in simulation and applied probability and their applications in a variety of fields. His work has spanned simulation and gradient estimation, applied probability, production-inventory systems

and operations management, and financial engineering and risk management. He has authored more than 110 journal articles and 11 book chapters. He has written three books and edited two more. Professor Glasserman’s first book, “Gradient Estimation via Perturbation Analysis,” is a standard reference in simulation with more than 700 citations. His book “Monte Carlo Methods in Financial Engineering” is a standard reference not just within financial engineering circles, but also in simulation. The book, widely known among practitioners, was recently translated into Chinese, and has more than 3,700 citations. Professor Glasserman has striven to reach not just the research community, but also practitioners, as evidenced by his “research briefs” written for the Office of Financial Research in the Treasury Department, and by his article, co-authored with

Paul Glasserman, recipient of the Saul Gass Expository Writing Award.

Mike Giles for Risk Magazine, “Smoking Adjoints: Fast Monte Carlo Greeks.” Professor Glasserman’s written work is invariably lucid, precise, illuminating and persuasive. It is no wonder that his work is heavily cited and extremely influential, not just within academic circles, but also in practice. ORMS

Chang receives Undergraduate Operations Research Prize Joy Chang of the University of Michigan won the 2016 Undergraduate Operations Research Prize from INFORMS for her paper, “Car Sharing Fleet Location Design with Mixed Vehicle Types for CO2 Emission Reduction.” The award was presented at the INFORMS Annual Meeting in Nashville, Tenn., by Murat Kurt following a series of presentations. Pavithra Harsha chaired the selection committee. The competition is held each year to honor a student or group of students who conducted a significant applied project in operations research or management science, and/or original and important theoretical or applied research in operations research or management science, while enrolled as an undergraduate student. The prize includes a monetary award of $500 plus travel support to attend the INFORMS Annual Meeting.

Honorable mention honors went to Hari Bandi of MIT for the paper, “Regularized Linear Regression via Robust Optimization Lens” and to Cem Aydın, Alp Arıbal, Cansu Erol and Begum Tuglu of Koc University, Turkey, for their paper, “A Reformulation of the Appointment Scheduling Problem with Customer Choice Behavior and Multiple INFORMS President Ed Kaplan (left) and Committee Chair Customer Types.” Murat Kurt (right) congratulate Joy Chang (middle). Other finalists and their papers included: Zeynep try Data Analysis for Quality Assessment Yaprak Besik, Basak Erman, Deniz Berfin of Biological Samples”); Jacob Monroe of Karakoc, Yekta Jehat Mizrakli, Umut Muru- North Carolina State University (“Allocating roglu and Egehan Yanik of Bilkent Univer- Countermeasures to Defend Water Distrisity, Turkey (“Hot Sales Delivery Logistics bution Systems Against Terrorist Attack”); Optimization for ET”); Sameer Manchada and Hale Erkan of Bilkent University, Turand Mikaela Meyer of Purdue University key (“Collaborative Decision Making for Air (“On Comprehensive Mass SpectromeTraffic Management”). ORMS December 2016

ORMS Today

| 49

n ews Reiman, Williams share von Neumann Prize The John von Neumann Theory Prize of INFORMS was awarded to Marty I. Reiman and Ruth J. Williams for the “seminal research contributions [they] have made over the past several decades to the theory and applications of stochastic networks/systems and their heavy traffic approximations.” Prize Committee Chair David Shmoys made the presentation at the INFORMS Award Ceremony held in conjunction with the 2016 INFORMS Annual Meeting in Nashville, Tenn. The prize recognizes scholars who have made fundamental, sustained contributions to theory in operations research and the management sciences. The citation read in part: [Reiman and Williams’] profound contributions have … led to breakthroughs in stochastic operations research in general, and queueing theory in particular. Their analysis of complex stochastic networks under conditions of heavy traffic has not only led to the discovery and rigorous articulations of properties of the networks, and penetrating insights into the operational laws of real-world systems they model, but also led to deep theoretical developments in the study of reflected diffusions. Starting with his Ph.D. thesis, Marty Reiman has had a lasting impact on the heavy traffic analysis of queueing systems. In it, he identified and characterized the diffusion limit of a generalized Jackson queueing network in heavy traffic. … In companion and subsequent works, Reiman enunciated two heavy-traffic principles that rank among the most important and elegant contributions of heavy traffic theory: the snapshot principle, which relates waiting- and sojourn-times processes to queue-lengths, and the phenomenon of statespace collapse, that is, dimensionality reduction in the effective descriptions of the evolution of a stochastic system in heavy traffic. Reiman’s research is characterized by deep intuition and penetrating understanding of the physical and mathematical laws that govern the systems that he studies. These virtues are clearly manifested in the following illustrative examples: the averaging principle for polling systems, jointly with Coffman and Puhalskii; the interpolation approximation, with 50 | ORMS Today

December 2016

Simon, that combines their lighttraffic and the familiar heavy-traffic viewpoints; the study, motivated by call centers, of operational regimes (efficiency-driven, quality-driven and their balancing QED, namely, quality-and-efficiency driven) in many-server queueing models with abandoning customers Committee Chair David Shmoys (far left) and INFORMS (with Garnett and Mandelbaum); President Ed Kaplan (far right) flank John von Neumann asymptotically optimal staffing of Prize recipients Ruth Williams and Marty Reiman. many-server queues (with Borst and Mandelbaum), e.g., square-root staffing discipline. Jointly, first with Gromoll and Puha in the Halfin-Whitt (QED) regime; the analysis and later with Puha and Stolyar, Williams dein the latter regime, with Puhalskii, of the scribed the evolution of such systems by a multiclass queue with phase-type servicemeasure-valued process that keeps track of times and static-priorities; the constant-order the residual service times of all jobs. Another policy in lost-sales inventory systems with long example is asymptotically optimal control of lead times; and, jointly with Wang, his analysis parallel-server systems in heavy-traffic, first of certainty equivalent control for network under complete resource pooling (with Bell) revenue management. and recently under partial pooling (with Pesic); In Reiman’s work, one sees real inven- further examples include applications, with tiveness combined with strong mathemat- Kelly, to the analysis of a controlled motorway ical and expository skills, supported by a in heavy traffic; and, with Kang, Lee and Kelly solid command of several distinct application and later with Gromoll, to bandwidth sharing domains. His research has influenced and networks that model congestion of data on inspired work by the very best people in sto- the Internet. For the latter, Williams estabchastic O.R., including several previous win- lished fluid limits and subsequently discovered ners of the von Neumann Theory Prize. that networks with proportional fairness admit Ruth Williams has also had a deep a product-form (and hence tractable) stationand lasting impact on the study of heavy traffic ary distribution in heavy traffic. analysis. This started with her Ph.D. thesis and Williams’ research is characterized by its further work on RBM in a wedge, establishing mathematical depth and elegance. She has semimartingale, local time, excursion and greatly influenced researchers in operations recurrence properties. It continued with her research, stochastic processes and mathestablishing a multidimensional generalization ematics, doing so through survey lectures that identified necessary and sufficient and articles that are exemplary in clarity and conditions for weak existence and uniqueness in insight. Her expositions have introduced the law of RBMs in the nonnegative orthant, which field to researchers and described challenging provides an alternative to the pathwise theory open problems and directions, which have of Harrison and Reiman; and it culminated spurred further research. in the development of invariance principles In summary, Reiman and Williams have (analogous to the Donsker-Varadhan invariance carried out pioneering research over the past principle for the classical (unreflected) Brownian several decades. This has led to fundamental motion). This enabled the identification and breakthroughs in stochastic operations verification of diffusion approximations, for research in general, and queueing theory in multiclass queueing networks under certain particular, with a focus on stochastic networks state-space collapse conditions. and their behavior under heavy-traffic The above research provided the foun- conditions. Their research, in which they have dations for subsequent significant research, by both influenced and built upon each other’s Williams and others. One example is queues work, has had a lasting theoretical and practical operating under a processor-sharing service impact that stands the test of time. ORMS ormstoday.informs.org

ormstoday.informs.org

Raghavan receives Teaching of OR/MS Practice Prize The 2016 INFORMS Prize for the Teaching of OR/MS Practice was presented to S. R aghu R aghavan of the University of Maryland for his “his professionalism and contributions to the teaching of practice in operations research and the management sciences.” Committee Chair Eli Olinick announced the award at the INFORMS Annual Meeting in Nasvhille, Tenn. A professor of management science and operations management at UMD’s Robert H. Smith School of Business, Raghavan was

recognized for having “succeeded in helping his students acquire the knowledge and skills necessary to be effective practitioners of operations research or the management sciences.” His research interests and activities cover a broad domain including auction design, data mining, economics, information systems, computational marketing, networks, optimization and telecommunications. ORMS

Committee Chair Eli Olinick (left) and INFORMS President Ed Kaplan (right) flank award-winning teacher S. Raghu Raghavan.

O’Mahony captures Dantzig Dissertation Award Eoin O’Mahony (Cornell University, now at Uber) was presented the 2016 George B. Dantzig Dissertation Award from INFORMS for the dissertation, “Smarter Tools for (Citi) Bike Sharing.” Committee chair Harry Groenevelt presented the award at INFORMS Annual Meeting in Nashville, Tenn. The award honors the best dissertation in any area of operations research and the management sciences that is innovative and relevant to practice. The award is designed to encourage academic research that combines theory and practice and stimulates greater interaction between doctoral students (and their advisors) and the world of practice.

Second place went to Pooyan Kazemian (University of Michigan, now at Massachusetts General Hospital and Harvard Medical School) for “Stochastic Control and Optimization for Chronic Disease Monitoring and Control, Hospital Staffing and Surgery Scheduling.” Honorable mention went Harry Groenevelt (left) and INFORMS President Ed Kaplan to Tugce Martagan (University (right) flank Eoin O’Mahony. of Wisconsin-Madison, now at Eindhoven University) for “Stochastic Models Nathaniel Bastian (Penn State University, to Optimize Biomanufacturing Operations,” now at U.S. Army Human Resources Vamsi Kanuri (University of Missouri, now Command) for “Multiple Criteria Decision at the University of Miami) for “Optimizing Engineering to Support Management a Menu of Multi-Format Subscription Plans in Military Healthcare and Logistics for Ad-Supported Media Platforms,” and Operations.” ORMS

Nicholson Student Paper Competition Hamsa Bastani of Stanford University won the 2016 George E. Nicholson Student Paper Competition for the paper, “Online Decision-Making with High-Dimensional Covariates.” The competition is held each year to honor outstanding papers in the field of operations research and the management sciences written by a student. Mehdi Behroozi of the University of Minnesota placed second with his paper, “Household-level Economies of Scale in

Transportation,” while Yun Zeng of The Ohio State University received honorable mention for the paper, “A Necessary and Sufficient Condition for Throughput Scalability of Fork and Join Networks with Blocking.” Finalists included Martin Zubeldia of MIT, Ali Aouad of MIT, Rui Gao of Georgia Tech and Rajan Udwani of MIT. The awards were presented by Committee Chairperson Maria Mayorga at the INFORMS Annual Meeting in Nashville, Tenn. ORMS

Hamsa Bastani (middle) is joined by Maria Mayorga (left) and INFORMS President Ed Kaplan (right). December 2016

ORMS Today

| 51

n ews Doing Good with Good O.R. Student Paper Competition Christine Barnett and Selin Merdan of the University of Michigan won the 2016 Doing Good with Good O.R. Student Paper Competition for the paper “Data Analytics for Optimal Detection of Metastatic Prostate Cancer.” The competition is held each year to identify and honor outstanding projects in the field of operations research and the management sciences conducted by a student or student group that have a significant societal impact. Finalists presented their papers in sessions held in conjunction with the INFORMS Annual Meeting in Nashville, Tenn. Ohad Eisenhandler and Michal Tzur of Tel Aviv University received honorable mention for their paper, “The Humanitarian Pickup and Distribution

Problem.” Fi n a l i s t s i n c l u d e d : Pantelis Loupos and Can Urguny of Northwestern U n i v e r s i t y ; S a i t Tu n c , Oguzhan Alagoz and Elizabeth S. Burnside of the University of WisconsinMadison; Semih Boz, Semih Kaldirim, Bilge Kaycioglu, Buse Eylul Oruc, Eren Ozbay, and Mirel Yavuz of Bilkent University; and Doing good (l-r): INFORMS President Ed Kaplan, Sinan Derindere, Ali Erkan competition winners Selin Merdan and Christine Barnett and Pinar Ozkurt of Turkish and committee co-chairs Karen Smilowitz and Chase Red Crescent, Strategic Rainwater. Management Department. Committee co-chairs Karen Smilowitz at the 2016 INFORMS Annual Meeting in and Chase Rainwater presented the awards Nashville, Tenn. ORMS

Volunteer Service Award New for 2016, the Volunteer Service Award recognizes up to 20 INFORMS members annually who have been engaged in volunteer service during the past year and whose contributions have had an impact in the area they serve. Volunteer Service Awardees must have at least one year of involvement in INFORMS and must not be current or incoming Board members. They are awarded at three levels: Service, Meritorious Service or Distinguished Service. Committee Chair Lauren Davis announced the inaugural Volunteer Service Award winners at the 2016 INFORMS Annual Meeting in Nashville, Tenn. The recipients: Distinguished Service • Dionne Aleman, University of Toronto • Theodore T. Allen, The Ohio State University • Wedad Elmaghraby, University of Maryland, RH Smith School of Business • Terry P. Harrison, CAP, Penn State University • Stefan E. Karisch, Boeing • Don N. Kleinmuntz, University of Notre Dame and Strata Decision Technology 52 | ORMS Today

December 2016

Committee Chair Lauren Davis (far left) and INFORMS President Ed Kaplan (third from right) congratulate several of the 2016 Volunteer Service Award recipients.

• Anna Nagurney, University of Massachusetts-Amherst, Isenberg School of Management • Douglas Samuelson, InfoLogix Meritorious Service • Deepak Agrawal, Penn State University • Peter C. Bell, University of Western Ontario, Richard Ivey School of Business • Aaron Burciaga, CAP, Accenture • Esra Buyuktahtakin, Wichita State University

• Carri Chan, Columbia University • Gino J. Lim, University of Houston • Ranganath S. Nuggehalli, CAP, United Parcel Service • Aurelie Thiele, Southern Methodist University Service • Karen T. Hicklin, North Carolina State University • Simge Küçükyavuz, Ohio State University • Doug M. Matty, U.S. Army • Daniel Reich, Ford Motor Company ORMS ormstoday.informs.org

ormstoday.informs.org

People Former MIT graduate student Yi Xue and MIT professor and former INFORMS President Richard Larson recently received the Lawrence R. Klein Award from the U.S. Bureau of Labor Statistics for the best article among nonBLS authors that appeared in the BLS publication Monthly Labor Review. Published in 2015, the prize-winning paper, “STEM crisis or STEM surplus? Yes and yes,” uses BLS data and other data to

show that, nationally, there is both a STEM crisis and a STEM surplus, depending on the particular employment field. One area of a shortage crisis is in computer sciencerelated fields, whereas one area of surplus is Ph.D.s in the life sciences. According to BLS sources, the Xue-Larson paper – available freely online – remains one of its most widely read papers, even nearly 20 months after first publication. The paper has been cited in national media including recently in The Wall Street Journal. From 2012 to 2014 Xue was a master’s student at MIT, working on NIH-supported research with Larson. Larson is co-principle investigator on a four-year NIH grant, “Developing a Scientific Workforce Analysis and Modeling Framework.” The award was established in honor of Nobel-Prize winner Lawrence R. Klein, who earned his Ph.D. in economics at Richard Larson (left) receives the Lawrence R. Klein MIT in 1944, where he was Award from Deborah Klein at an awards ceremony at the BLS in Washington, D.C. Paul Samuelson’s first doctoral

student. Dr. Klein retired in 1968 after 22 years as editor in chief of the Review and established a fund to encourage articles that exhibit originality of ideas or methods on analysis, adhere to the principles of scientific inquiry and are well written. Michael J. Armstrong has been awarded a $17,000 research fellowship from Australia’s Endeavour program. The fellowship will support him as a visiting researcher at the University of New South Wales campus in Canberra, Australia, for four months in 2017. He will be working on military O.R. projects in collaboration with professors from the School of Humanities & Social Sciences. Armstrong is an Michael J. associate professor of Armstrong operations research in the Goodman School of Business at Brock University in St. Catharines, Ontario, Canada. ORMS

In Memoriam

Marius M. Solomon, 1955-2016 Marius M. Solomon, a longtime professor at Northeastern University, an INFORMS Fellow and a well-known thought leader in the management science field, passed away Sept. 28 at the age of 61. A professor of supply chain and information management in the D’AmoreMcKim School of Business at Northeastern University in Boston where he spent his entire professional career, Dr. Solomon served INFORMS in several capacities, including president of the INFORMS Transportation Science and Logistics Section and associate editor of the INFORMS journals Operations Research and Transportation Science. The group coordinator for the Supply Chain and Information Management Group at Northeastern from 2008-2014, Professor Solomon taught operations and supply

chain courses for undergraduate, graduate and executive students from 1983-2016. Born on Feb. 2, 1955, in Bucharest, Romania, Dr. Solomon received his Ph.D. in decision sciences from the University of Pennsylvania’s Wharton School of Business. He also earned a master’s degree in dynamical systems and a bachelor’s degree in mathematics from the University of Pennsylvania. Professor Solomon’s research and consulting interests focused on the development and implementation of design, planning and execution methodologies leading to improved supply chain agility. He made major contributions to the development of optimization approaches for time sensitive aspects of supply chains that benefit the public and private sectors. He published numerous articles addressing time constrained vehicle fleet planning, routing

and crew scheduling, supply chain design, and the management of advanced manufacturing and warehousing systems. A worldwide traveler, Professor Solomon is remembered by his friends and colleagues for “recounting his adventures during lunches and dinners, filling rooms with laughter and joy. He was a great friend, a respected colleague, and a passionate teacher, and he will be dearly missed by all of us in the Northeastern University community.” Professor Solomon is survived by his daughters Hannah N. Solomon and Arielle M. Solomon of Milton, Mass. ORMS Sources: Northeastern University, The Boston Globe, INFORMS December 2016

ORMS Today

| 53

n ews

Scenes from

Nashville 2016 Image ÂŠ Nashville Convention & Visitors Corporation

Clockwise, from right: The Nashville skyline provided a picturesque backdrop for the 2016 INFORMS Annual Meeting; Music City was swinging; awards galore; the plenary presentations were crowd-pleasers; poster session action; old friends Al Blumstein and Randy Robinson enjoying the meeting; and three young friends doing the same.

54 | ORMS Today

December 2016

ormstoday.informs.org

Clockwise, from above: Good eats at the general reception; hats and a handful of hors dâ&#x20AC;&#x2122;oeuvres go great together; INFORMS staffers Nagaraj Reddi, Kara Tucker, Olivia Schmitz and Ashley Kilgore; Music City rocks; and rolls; and rocks some more; heavy foot traffic for exhibitors; and get ready to do it again next year when the 2017 INFORMS Annual Meeting heads for Houston.

December 2016

ORMS Today

| 55

n ews

Meetings Conference snapshot survey:

Analytics student population continues to grow The population of students studying analytics is increasing, according to an informal Princeton Consultants snapshot survey at the INFORMS Annual Meeting from Nov. 13-16 in Nashville. For most students their future plans are academic or research-oriented; for those who intend to become practitioners, consulting is the leading industry choice. Irv Lustig, Princeton Consultants Optimization Principal and longtime active INFORMS member, reports the following findings: • 23.6 percent of students said they intend to become analytics practitioners at for-profit or non-profit organizations; the remainder plan to remain in academia or perform research. • For students intending to work at a forprofit business, consulting is the most preferred of 14 industries, followed by

financial services and healthcare. • 94 percent of professors said the student population in analytics at their schools has increased or remained the same size during the last three years. • Professors who belong to INFORMS teach analytics in a variety of departments and schools, led by Business Administration (MBA) and the Operations Research Department. The survey included 111 self-selected participants, all of whom were onsite at the conference. The participants included 73 students, 33 professors and five practitioners. The participating professors were asked the following question: “In the past 3 years at my institution, the size of the student population that is studying O.R./ analytics has …” The results are shown in the following table:

INFORMS Annual & International Meetings April 2-4, 2017 INFORMS Conference on Business Analytics & Operations Research Caesars Palace, Las Vegas Las Vegas, Nevada Chair: Maher Lahmar, IBM http://meetings2.informs.org/analytics2017/

July 26-28, 2017 INFORMS 2017 Healthcare Conference

De Doelen International Congress Centre Rotterdam, the Netherlands Chair: Joris van de Klundert, Erasmus University Rotterdam http://meetings2.informs.org/healthcare2017/

Oct. 22-25, 2017 INFORMS Annual Meeting

George R. Brown Convention Center & Hilton Americas Houston, Texas Chair: William Klimack, Chevron

2018

April 15-17, 2018 INFORMS Conference on Business Analytics & Operations Research Marriott Waterfront Hotel, Baltimore

INFORMS Community Meetings Jan. 15-17, 2017 INFORMS Computing Science Conference Westin Austin at the Domain Austin, Texas Chair: Neil Dimitrov, UT- Austin https://ie.clemson.edu/ics2017/

Feb. 2-5, 2017 INFORMS Organization Science Conference Grand Summit Hotel, Canyons Resort Park City, Utah http://pubsonline.informs.org/page/orsc/winterconf23

June 7-10, 2017 INFORMS Marketing Science Conference University of Southern California, Los Angeles Los Angeles Chair: Gerry Tellis https://marketingscience2017.usc.edu

June 26-27, 2017 INFORMS Advances in Decision Analysis Call for Nominations, from p. 45

The president-elect will serve three years: one year as president-elect, one year as president and one year as past president. INFORMS elections begin in the summer, in recent years running from August through the end of September. INFORMS employs a voting process called approval voting. This means that INFORMS members can vote for as many candidates as they think are qualified for a position. According to Bylaw 3: Elections shall be conducted by approval voting, whereby each voter may vote for any number of candidates for an office and the individual elected shall be the

56 | ORMS Today

December 2016

one receiving the largest number of votes … A quorum for membership voting is 10 percent of the members eligible to vote. If you think that someone, including yourself, would be a good candidate for one of the aforementioned openings, please consider sending your recommendation(s) to nominations@ informs.org. Nominations received by Jan. 10, 2017, will be considered. Nominations received after that date may be considered. The nominating committee is scheduled to report its nominations by Feb. 15, 2017. ORMS

The University of Texas at Austin Austin, Texas Chair: Casey Lichtendahl, University of Virginia https://www.informs.org/Community/DAS/ADA-2017Conference

June 29-30, 2017 INFORMS Revenue Management & Pricing Section Conference

Centrum Wiskunde & Informatica (CWI) Amsterdam Chair: Arnoud Den Boer https://www.informs.org/Community/revenue-mgt/Conferences

July 10-12, 2017 INFORMS 19th Applied Probability Conference Northwestern University Evanston, Ill.

Go to www.informs.org/Conf for a searchable INFORMS Conference Calendar.

ormstoday.informs.org

CROP CHALLENGE in ANALYTICS

The 2017 Syngenta Crop Challenge in Analytics focuses on the seed retailer, who sells soybean seed varieties to farmers. The farmers require different soybean seed varieties based on expected growing conditions. To maximize yield in a region with a large number of farmers, the retailers need to predict and stock the soybean variety seeds that will thrive best in the farmer’s most common growing conditions. It is difficult for a seed retailer to predict which seed varieties to stock almost a year in advance to the soybean crop planting by the farmers.

THE CHALLENGE:

Which soybean seed variety, or mix of up to five varieties in appropriate proportions, will best meet the demands of farmers in a growing region?

CHALLENGE LAUNCH - Data Available NOW! • Deadline for submissions January 16, 2017 • Selection of finalists February 24, 2017 • Finalist presentation (live or via telepresence) at INFORMS Conference on Business Analytics and O.R. April 2–4, 2017

Industry News

Frontline surpasses 200,000 cloudbased advanced analytics users Frontline Systems, developer of the Solver in desktop Microsoft Excel 26 years ago, announced that it has surpassed 200,000 users of its cloud-based advanced analytics tools for optimization, simulation/risk analysis, forecasting, data mining and text mining – based on usage data from Microsoft, Google and its own SaaS platforms. “We moved early to the cloud with Solver for Excel Online in July 2013,” says Daniel Fylstra, Frontline’s founder and president. “In 2016, we’re seeing largescale adoption of all our cloud advanced analytics tools.” Usage has doubled from 100,000 users in January to 200,000 users today. Frontline has been known for many years as a leading developer of advanced analytics add-ins for desktop Excel. It has also offered analytics tools in other forms, such as SDKs for developers, since the mid-1990s. Today, Frontline offers seven advanced analytics tools for Excel Online and Google Sheets, two selfcontained cloud-based SaaS platforms – XLMiner.com and AnalyticSolver.com – and RASON (RESTful Analytic Solver Object Notation), its analytics modeling language and REST API. It also offers Analytic Solver Platform for desktop Excel, and Solver SDK and XLMiner SDK for developers. AnalyticSolver.com, Frontline’s new, integrated cloud SaaS platform for predictive and prescriptive analytics, was launched two months ago, but Frontline has already granted requests for access to the platform to business students in more than 450 university courses, as well as two currently running massive open online courses (MOOCs) – one on U.S.-based Coursera and one in China and India – and its own online education platform Solver Academy, based on Open edX. Frontline offers both desktop and cloud versions, with a very high level of interoperability between them. Frontline also offers deep functionality and high performance in the areas of data mining and predictive analytics, mathematical optimization, and simulation and risk 58 | ORMS Today

December 2016

analysis tools. For more information, visit www.solver.com. Fast-growing AIMMS earns FD Gazelle Award AIMMS recently earned an FD Gazelle Award, given annually to the fastest growing companies in the Netherlands. “We are extremely proud to win an FD Gazelle Award,” says Gijs Dullaert, CEO of AIMMS. “It is a great acknowledgement for our valuable customers and our growing team.This also confirms that our mission to bring the benefits of prescriptive analytics to society is the right path forward, and we will continue to fulfill it with commitment and passion.” Het Financieele Dagblad (the Dutch Financial Times) organizes the FD Gazelle Awards. Criteria that play a role in the selection include that the company is financially healthy, has a minimum turnover of 100,000 Euros and a revenue growth of at least 20 percent in the past three financial years. For more information, visit: www. aimms.com. Gurobi introduces Gurobi Optimizer v7.0 Gurobi Optimization recently introduced Gurobi Optimizer v7.0, with higher performance and powerful new modeling capabilities. Some of the enhancements and new features include: • Significant performance improvements across MIP, LP, QCP, MIQCP and MIQP problem types. • Python modeling enhancements, including new methods and classes that further simplify the task of translating mathematical models into efficient implementations. • Support for multiple objectives, allowing users to associate multiple, independent optimization objectives with their model and then perform either blended or lexicographic optimization on the resulting multiobjective model. • MIP solution pool support, so users can obtain more than just one optimal solution to an MIP model.

• General constraints, so users can enter commonly occurring constraints (MIN/MAX, ABS, AND/OR and Indicator constraints) without having to translate them into linear constraints. • Support for tuning criterion, so users can search for settings that produce the best lower bound, upper bound or gap. In addition, Gurobi has added support for Python 3.5 on the Mac, enhanced its .NET property support and added a number of useful new parameters. For more information, visit http:// www.gurobi.com/. Credit unions embrace FICO Score Open Accesss U.S. financial institutions of all sizes are offering the FICO Score Open Access program to their customers as a way to empower them to take control of their financial health. Participants in the program provide FICO Scores and related information to their customers, and participating lenders range from the country’s largest lenders to dozens of credit unions, some with as few as 300 members. Thanks to these lenders, 180 million consumer accounts are now getting their FICO Scores for free. “We’re ver y satisfied with the comprehensive education including the FICO Score, key factors that affected the score, and score-related frequently asked questions that our members receive as part of this program,” said Amy ForcierPabst, EVP of Member Experience at Royal Credit Union, which services Wisconsin and Minnesota. “We anticipate increased online engagement from our members as a result of this program.” The FICO Score is the standard measure of U.S. consumer credit risk, and is used in more than 20 countries. More than 10 billion FICO Scores are purchased in the United States each year by lenders for their risk management decisions.Through its myFICO service and the FICO Score Open Access program, FICO helps millions of American consumers understand their credit health. ORMS ormstoday.informs.org

SPECIAL ADVERTISING SECTION | View Classifieds Online at: http://www.orms-today.org

CLASSIFIEDS

THE UNIVERSITY OF TEXAS AT DALLAS The Naveen Jindal School of Management

INFORMATION SYSTEMS TENURE-TRACK FACULTY POSITION

The Naveen Jindal School of Management at the University of Texas at Dallas VHHNV WR Ă&#x20AC;OO RQH RU WZR WHQXUH WUDFN IDFXOW\ position(s) at the Assistant Professor rank in the Information 6\VWHPV DUHD 7KH DSSRLQWPHQWV ZLOO EH HIIHFWLYH )DOO &DQGLGDWHV ZLOO EH UHTXLUHG WR FRQGXFW KLJK TXDOLW\ UHVHDUFK DQG SXEOLVK KLV KHU UHVHDUFK LQ WRS MRXUQDOV LQ WKH LQIRUPDWLRQ V\VWHPV DUHD GHYHORS DQG WHDFK EXVLQHVV DQDO\WLFV FRXUVHV LQ WKH XQGHUJUDGXDWH DQG JUDGXDWH SURJUDPV DQG HQJDJH LQ VHUYLFH DFWLYLWLHV ZLWKLQ WKH XQLYHUVLW\ DQG WKH SURIHVVLRQ $ 'RFWRUDO GHJUHH LV UHTXLUHG SULRU WR MRLQLQJ EXW WKRVH ZLWK $%' VWDWXV ZLOO EH FRQVLGHUHG DW WKH DSSOLFDWLRQ LQWHUYLHZLQJ VWDJH

5HYLHZ RI DSSOLFDWLRQV ZLOO EHJLQ LPPHGLDWHO\ DQG FRQWLQXH XQWLO WKH SRVLWLRQ V DUH Ă&#x20AC;OOHG

$SSOLFDQWV VKRXOG VXEPLW WKHLU FXUULFXOXP YLWDH HGXFDWLRQDO KLVWRU\ WHDFKLQJ UHVHDUFK DQG RWKHU SHUWLQHQW ZRUN H[SHULHQFH OLVW RI SXEOLFDWLRQV DQG SUHVHQWDWLRQV DQG three reference letters via the online application system at https://jobs.utdallas.edu/postings/6618. 7KH 8QLYHUVLW\ RI 7H[DV DW 'DOODV LV DQ (TXDO 2SSRUWXQLW\ $IĂ&#x20AC;UPDWLYH $FWLRQ HPSOR\HU DQG VWURQJO\ HQFRXUDJHV DSSOLFDWLRQV IURP FDQGLGDWHV ZKR ZRXOG HQKDQFH WKH GLYHUVLW\ RI WKH 8QLYHUVLW\ÂˇV IDFXOW\ DQG DGPLQLVWUDWLRQ

CHAIR, DEPT. OF INDUSTRIAL AND OPERATIONS ENGINEERING, University of Michigan The Department of Industrial and Operations Engineering at the University of Michigan seeks applicants and nominations for the position of Department Chair. The Department currently has 31 full-time faculty members with approximately 550 undergraduate and 230 graduate students. Graduate education leading to M.S., M.S.E., and Ph.D. degrees is conducted in a wide variety of topic areas, including all areas of Operations Research, Analytics, Ergonomics & Human Factors, Risk Management, and Quality and Applied Statistics. Application areas include Manufacturing, Healthcare Operations and Medical Decision Making, Energy, Risk Analysis, Supply Chain Management, Service Systems, Sustainability, and Transportation. The successful candidate will be an outstanding scholar with an earned doctorate in a research field related to Industrial and Operations Engineering and will have an exemplary record of achievement in research, teaching and service at a level commensurate with appointment as a tenured full professor. He or she must also possess visionary leadership abilities, a broad appreciation for the diverse perspectives within Industrial and Operations Engineering, and a strong interest in promoting sponsored research programs and mentoring faculty. The qualified candidate should be able to lead and support the faculty to ensure that learning of the highest quality flourishes at all levels, from undergraduate education to graduate and post-doctoral research. The candidate should be able to work with a diverse group of faculty, staff, students, and administrators to achieve common goals and to maintain rapport with alumni and industry representatives. The University of Michigan is a non-discriminatory/affirmative action employer. Underrepresented minorities and women are strongly encouraged to apply. The College of Engineering is especially interested in qualified candidates who can contribute, through their research, teaching, and/or service, to the diversity and excellence of the academic community and who will build collaborative ties with other departments within the College of Engineering and the University. The University of Michigan is responsive to the needs of dual career families. Applicants should electronically submit a detailed curriculum vitae and cover letter describing professional background, qualifications, and leadership experience as well as a two-page synopsis of their views on the current challenges and opportunities facing industrial and operations engineering education and research. The deadline for ensuring full consideration of an application is December 1, 2016, but the position will remain open and applications may still be considered until the appointment is made. The search will be conducted in confidence until finalists are invited for campus visits at which time professional references will be contacted. Please submit your application to the following: Web: http://www.engin.umich.edu/ioe/careers/chair-search If you have any questions regarding the web application submittal process or other inquiries please contact Professor Lawrence M. Seiford, Chair, IOE Search Committe, at seiford@umich.edu.

ARIZONA STATE UNIVERSITY FACULTY POSITIONS The Ira A. Fulton Schools of Engineering at Arizona State University (ASU) seek outstanding applicants for tenure track/tenured faculty positions. Active searches are being conducted in the following areas but the excellence of the candidateâ&#x20AC;&#x2122;s accomplishments and potential are more important than the specific area: Advanced Industrial and Manufacturing Engineering Systems (Job #11777) Areas of interest include: production control and manufacturing management; analytics for next-generation manufacturing systems; process capability, optimization and reliability; advanced processes and systems for product design; automation; and manufacturing enterprise systems. Industrial Statistics/Stochastic Optimization (Job #11778) Areas of interest include: methodologies for big data, computational statistics, and applications in IoT, data analytics, image/text/voice processing, however the vision, leadership potential and record of accomplishments is a higher priority than the specific area of research. Appointments will be made at the rank commensurate with experience and accomplishments starting August 2017. A successful candidate will hold an earned doctorate in a relevant field and have demonstrated evidence of excellence in research and teaching. Desired qualifications include a record of external funding, publication in top tier journals, innovative pedagogy and participation and leadership in collaborative, transdisciplinary research with high societal impact as appropriate to the candidateâ&#x20AC;&#x2122;s rank. Although the faculty appointment may be in any of the Fulton Schools of Engineering, the Industrial Engineering program in the School of Computing, Informatics, and Decision Systems Engineering is the most involved in the interest areas of the search. Current information regarding these positions and instructions for applying are available at http://engineering.asu.edu/hiring/. Review of applications for each search will begin December 10, 2016; if not filled, reviews will occur the 1st and 15th of every month there after until the search is closed. Arizona State University is a VEVRAA Federal Contractor and an Equal Opportunity/Affirmative Action Employer. All qualified applicants will be considered without regard to race, color, sex, religion, national origin, disability, protected veteran status, or any other basis protected by law. See ASUâ&#x20AC;&#x2122;s full non-discrimination statement (ACD 401) at https://www.asu.edu/aad/manuals/acd/acd401.html and the Title IX statement at https://www.asu.edu/titleIX/.

December 2016

ORMS Today

| 59

CLASSIFIEDS

SPECIAL ADVERTISING SECTION | View Classifieds Online at: http://www.orms-today.org

PROFESSOR AND CHAIR

Industrial and Systems Engineering (ISE) Herbert Wertheim College of Engineering, University of Florida Applications and nominations are invited for the position of Professor and Chair of the Department of Industrial and Systems Engineering (ISE) in the Herbert Wertheim College of Engineering at the University of Florida (UF), the flagship campus of the State of Florida university system. The ISE Department offers B.S., M.S., and Ph.D. degree programs with an enrollment of about 500 full-time undergraduate students and 165 graduate students. The ISE Department has 11 tenured or tenure-track faculty members with several INFORMS, IISE, AAAS and IEEE Fellows. The Department also has 3 NSF CAREER Award winners and 1 ONR YIP Award winners. The Department’s current external research expenditures exceed $2 million annually. Areas of research strength in the Department include: Applied Deterministic and Stochastic Optimization, Network Optimization, Transportation, Financial, Supply Chain and Health Care Applications. For more information about the Department of ISE and the Herbert Wertheim College of Engineering, please visit www.ise.ufl.edu and www.eng.ufl.edu, respectively. In addition to overseeing the operational management of the Department, responsibilities of the Chair include: (1) creating a compelling vision for the advancement of the department to include increasing existing strengths, fostering new disciplines in emerging fields within ISE, and strengthening interdisciplinary efforts across the College and University; (2) facilitating both the professional and scholarly growth of the faculty, particularly the junior faculty; (3) ensuring cutting-edge education is provided to all trainees (undergraduate, graduate, postdoctoral and researchers); (4) enhancing the working partnership with the leaders of UF colleges and departments, administrations, industry and government agencies along with facilitating knowledge and technology transfer with industry; (5) recruiting a diverse faculty and student body; (6) increasing sponsored research, private and external funding opportunities for the department, cultivate corporate, governmental, alumni and other private donations. Qualifications: Earned Ph.D. and research and teaching experience in industrial engineering, operations research or closely related field is required. In addition, we are seeking an individual who is a distinguished scholar in their field of research, with demonstrated academic credentials sufficient for appointment at the Full Professor level; is committed to high academic standards; is skilled in the development and expansion of sponsored research programs; is experienced in enhancing the representation and success of underrepresented populations; has excellent leadership, management and interpersonal skills as well as written and oral communication skills; and encourages open, collaborative and inclusive problem solving. All candidates should apply through the UF Jobs website: https://jobs.ufl.edu/, reference position 494995. The Search Committee will begin reviewing applications on December 1, 2016, and continue accepting applications until the position is filled. For further questions, you may contact the search committee chair Dr. John Harris at harris@ece.ufl.edu. The University of Florida is an Affirmative Action/Equal Opportunity Employer; women, minorities and other under-represented groups are encouraged to apply.

SENIOR INFORMATION MANAGEMENT (HEALTHCARE) FACULTY POSITION Department of Information, Risk, and Operations Management McCombs School of Business The University of Texas at Austin The McCombs School of Business at the University of Texas at Austin invites applications for a potential senior faculty position as an Associate or Full Professor in Information Management, starting in Fall 2017. Candidates for this senior position must have a strong record of research in healthcare information management, and be able to teach related courses in our MBA and undergraduate programs. Applicants should have a Ph.D. in Information Management (or related disciplines). As an Equal Opportunity Employer with a commitment to diversity, we want our applicant pool to be as diverse as possible. We welcome applicants from under-represented groups as well as applicants who have demonstrated, and will continue to demonstrate, a commitment to diversity in the academic environment. The McCombs School, with its top-ranked faculty and educational programs, offers a stimulating and collegial environment for research and teaching. The Information Management group, which is part of the Information, Risk, and Operations Management (IROM) Department, is ranked among the best in the nation, and offers information management concentrations in the undergraduate, MBA, and PhD programs. The school’s research centers, including centers for electronic commerce, entrepreneurship, marketing, supply chain management and energy management, provide opportunities to collaborate with colleagues in other disciplines and interact with industry. Interested applicants should upload a curriculum vitae, three letters of reference, copies of research papers, course evaluation summaries, and a statement of research objectives and accomplishments to date to https://apply.interfolio.com/38379. Applications and recommendation letters must be received by December 4, 2016. To be considered for an interview at the INFORMS Nashville meeting, kindly upload application materials by November 1, 2016 (reference letters can be sent by December 4, 2016), and indicate the conference session(s) in which you will be presenting your work. The University of Texas at Austin is an Equal Opportunity Employer with a commitment to diversity at all levels. All qualified applicants will receive consideration for employment without regard to race, color, religion, gender, national origin, age, disability or veteran status. (Compliant with the new VEVRAA and Section 503 Rules.)

Professor of Professional Practice in the Department of Industrial Engineering and Operations Research The Department of Industrial Engineering and Operations Research (IEOR) at Columbia Engineering invites applications for a Professor of Professional Practice in Financial Engineering starting in the 2017-2018 academic year. Professor of Professional Practice are full-time non-tenure-track faculty members who have substantial professional expertise. The IEOR Department is looking for thought leaders with demonstrated scholarship and distinguished professional experience. Candidates for this position must hold a PhD in Operations Research, Financial Mathematics, or related disciplines. A successful candidate would help co-lead the educational mission of the Department in the field of Financial Engineering, broadly defined. This mission includes developing curriculum and courses relevant to the financial services industry, executive programs for practitioners, developing industry contacts for sponsored projects and project-based courses, and overseeing all aspects of the MS in Financial Engineering program, including admissions and student advising. The teaching and executive programs are expected to leverage blended and online learning environments. We expect the successful candidate to take a leadership role in the Center for Financial Engineering in the IEOR Department, and the Center for Financial and Business Analytics in the Data Science Institute by organizing seminars and conferences, and recruiting industry affiliates. The Department is especially interested in qualified candidates who can contribute, through their research, teaching, and/or service, to the diversity and excellence of the academic community. For additional information and to apply, please see: http://engineering.columbia.edu/faculty-job-opportunities. Applications should be submitted electronically and include the following: curriculum-vitae including a statement of teaching interests and plans, and contact information for three experts who can provide letters of recommendation. At least two of the letters of recommendation must address teaching ability. Candidates will be considered on a rolling basis with candidates applying before December 15, 2016 receiving full consideration. Applicants can consult www.ieor.columbia.edu for more information about the Department. Columbia University is an Equal Opportunity/Affirmative Action employer – Race/Gender/Disability/Veteran.

60 | ORMS Today

December 2016

ormstoday.informs.org

SPECIAL ADVERTISING SECTION | View Classifieds Online at: http://www.orms-today.org

CLASSIFIEDS

The Department of Systems Science and Industrial Engineering (SSIE) in the Thomas J. Watson School of Engineering and Applied Science at Binghamton University (State University of New York) is expanding further and seeks five (5) faculty positions starting in Spring or Fall 2017: • Assistant Professor – Smart Grids/Operations Research (Queueing Processes, Stochastic Optimization) – One (1) Position • Assistant Professor – Healthcare Systems Engineering – Two (2) Positions • Assistant Professor – Advanced Manufacturing (additive manufacturing; novel/advanced manufacturing systems/concepts) – Two (2) Positions One of the Healthcare Systems Engineering positions and one of the Advanced Manufacturing positions have a start date of Fall 2017. These positions require research that complements the Binghamton University Trans-disciplinary Areas of Excellence (TAEs: http://www.binghamton.edu/tae/). These positions involve establishing externally funded theoretical and applied research in the respective disciplines as well as teaching at all levels (undergraduate courses through advanced graduate courses). A reduced teaching load is granted during the first 2-3 years. The ideal candidates must have (i) an earned doctorate in industrial engineering or a related field, (ii) excellent leadership skills, and (iii) superb records of research, including garnering funding and scholarly publications. Teaching experience is preferred. The Watson School is dedicated to the goal of building a diverse and inclusive teaching, research, and working environment. Potential applicants who share this goal, especially underrepresented minorities, women, and persons with disabilities, are strongly encouraged to apply. The SSIE Department secures over 2.5 million dollars in research funding per year. Our faculty members secure research funding from over 25 sponsors from industry and federal agencies. The department currently has 20 faculty members and offers a BS degree in Industrial and Systems Engineering, and MS and PhD degrees in both Systems Science and in Industrial and Systems Engineering. SSIE also offers a cutting edge executive Master of Science in Health Systems Engineering program in Manhattan. With about 235 undergraduate, 230 master’s, and 120 doctoral students, the department is rapidly growing in numbers and in reputation. More details about the department are available at http://www.binghamton.edu/ssie/. Review of applications begins November 28, 2016, and continues until each position is filled. For more details about each position and to submit an application online, please visit http://binghamton.interviewexchange.com. Any questions about these positions should be directed to Dr. Mohammad T. Khasawneh, Professor and Department Chair (email:mkhasawn@binghamton.edu; phone: 607-777-4408; fax: 607-777-4094). About Binghamton University: Binghamton University is a highly ranked "Public Ivy." Binghamton University has built a reputation as a world-class institution that combines a broadly interdisciplinary, international education with one of the most vibrant research programs in the nation. Binghamton is proud to be ranked among the elite public universities in the nation for challenging our students academically, not financially. The result is a unique, best-of-both-worlds college experience. Our academic culture rivals a first-rate private university - rigorous, collaborative and boldly innovative -- while our campus culture exemplifies the best kind of public university experience: richly diverse students, active social life and deep engagement with the community. Located in the scenic Finger Lakes region of NY, Binghamton is three hours from New York City and Philadelphia, one hour from Ithaca and Syracuse, and five hours from Washington, DC and Boston. Binghamton is a low cost of living regional medical/hi-tech hub for 200,000 people. Binghamton University (The State University of New York) is an Equal Opportunity/Affirmative Action Employer. Women and Historically Underrepresented Minority Applicants are Strongly Encouraged to Apply.

December 2016

ORMS Today

| 61

CLASSIFIEDS

SPECIAL ADVERTISING SECTION | View Classifieds Online at: http://www.orms-today.org

FACULTY POSITIONS in The Grado Department of Industrial and Systems Engineering Virginia Tech The Grado Department of Industrial and Systems Engineering (ISE) at Virginia Tech invites applications for two tenure-track/tenured faculty positions effective August 2017, one at the rank of Assistant Professor and one at any rank, with preference for a senior faculty member. For particularly well-established Full Professor candidates with outstanding credentials, an endowed professorship is available. We seek outstanding candidates in all areas of Operations Research. Candidates will have the opportunity to work with a wide range of research groups and faculty within the Department, College, and University, including those working in the areas of data analytics and decision sciences, health systems and technology, and intelligent infrastructure, among others. The ISE Department is comprised of 30 full-time faculty with approximately 550 undergraduate students, 170 master’s students, and 90 doctoral students. The undergraduate and graduate ISE programs are currently ranked 5th and 9th, respectively, by U.S. News & World Report. Candidates are expected to lead innovative and high-quality research, build a strong sponsored-research program, develop and teach graduate and undergraduate courses, and advise and mentor graduate and undergraduate students. Candidates for the senior position should have a record of achievement commensurate with a senior faculty member, and are expected to provide organizational and research leadership, mentor junior faculty, and build collaborative relationships, both within the Department and across the College of Engineering and University. The position requires a Ph.D. in industrial and systems engineering, operations research, or a closely related field. Virginia Tech is committed to building a culturally diverse faculty and strongly encourages applications from women and minorities. Interested individuals should apply online at jobs.vt.edu (posting number TR0160134). Candidates should submit a cover letter, current CV, research statement, teaching statement, three relevant research publications, and the names of at least three references. Applicants interested in meeting with a faculty member at the 2016 INFORMS conference should contact the Search Committee at (ise-search@vt.edu). Applicants, particularly for the Assistant Professor position, are also encouraged to include in their cover letter a list of presentations being given at the 2016 INFORMS conference. Review of applications will begin immediately, and the deadline for ensuring full consideration is December 12, 2016. Positions will remain open, and applications may be considered until the position is filled. For more information or for any questions about the search, please contact the Search Committee at (ise-search@vt.edu). Virginia Tech does not discriminate against employees, students, or applicants for admission or employment on the basis of race, gender, disability, age, veteran status, national origin, religion, sexual orientation, or political affiliation. Virginia Tech is the recipient of a National Science Foundation ADVANCE Institutional Transformation Award to increase the participation of women in academic science and engineering careers. The ISE Department strongly supports the Virginia Tech Principles of Community. More information about the Department can be found at www.ise.vt.edu.

62 | ORMS Today

December 2016

FACULTY POSITIONS IN LYLE EMIS DEPARTMENT Southern Methodist University Open rank Faculty Positions Position Number 00005767 and 00052679 The Department of Engineering Management, Information, and Systems (EMIS) invites nominations and applications for an opportunity to be involved in the shaping of innovative academic programs in Operations Research, Management Science, Engineering Management, Systems Engineering, and Information Engineering. We seek outstanding candidates for open rank faculty positions in all areas relevant to our academic programs and all areas of industrial and systems engineering – methodological and applied – including but not limited to advanced data analytics, optimization, stochastic modeling, simulation, and model-based systems engineering with applications in supply-chain, manufacturing, health-care, in supply-chain, manufacturing, health-care, information, energy, and defense systems. Eligible candidates for the positions must have completed requirements for a doctoral degree in operations research, industrial engineering, systems engineering, or related field by August 2017; and must have the expertise to teach courses in areas relevant to our programs at the undergraduate, master’s, and doctoral levels. Eligible candidates for a tenure-track position (Position No. 00005767) are expected to demonstrate the ability to develop a strong, externally-funded research program and help advance the frontiers of knowledge. Candidates for Associate or Full Professor should have a commensurate record of research publications and external funding and an outstanding potential for research program development and research leadership. Candidates for Senior Lecturer or Professor of Practice (Position No. 00052679) should have relevant practical experience and are expected to contribute to both teaching and internal service missions of the department. Extraordinary candidates at all levels will be considered. SMU is a leading private university dedicated to academic excellence. Located near the center of Dallas, Texas, SMU enrolls 11,000 students, with nearly half in graduate programs. The EMIS department resides within the Bobby B. Lyle School of Engineering (http://www.smu.edu/lyle) founded in 1925 and offers a strong program of research and education at all levels, including Ph.D. degrees in operations research and systems engineering (http://www.smu.edu/Lyle/Departments/EMIS). The school provides an exceptional environment supporting multi-disciplinary collaborations and academic outreach and houses several institutes and centers -- with generous endowment support – relevant to research and teaching programs of the EMIS Department. These include the Hunter and Stephanie Hunt Institute for Engineering and Humanity, Darwin Deason Institute for Cyber Security, Caruth Institute for Engineering Education, and Hart Center for Engineering Leadership. SMU is designated as a preferred employer in the Dallas/Forth Worth (DFW) metroplex, one of the most prolific industrial centers in the country and a dynamic region with leading high-technology companies in the aerospace, defense, energy, information technology, life sciences, semiconductors, telecommunications, transportation, and biomedical industries. Some of the top companies and research institutes with a strong presence in the DFW area include Texas Instruments, Raytheon, Lockheed-Martin, Bell Helicopter, Frito-Lay, BNSF Railway, Turner Construction, Jacobs Engineering, Trinity Industries, Huitt-Zollars, Inc., The Beck Group, University of Texas Southwestern Medical Center, Parkland Health and Hospital System, and Baylor Research Institute. DFW is a multi-faceted community, offering exceptional museums, diverse cultural attractions and a vibrant economy. Dallas’ quality of life is exceptional with a relatively low cost of living, upscale apartments and homes within walking distance of SMU campus, the opportunity to live in the city or out in the country with a relatively short commute. To learn more about the rich cultural environment of SMU, please see: http://www.smu.edu. The target appointment date is the fall semester, 2017. To ensure full consideration for the position, the application must be emailed by November 21, 2016, but the committee will continue to accept applications until the position is filled. Interested and qualified applicants should email a curriculum vitae, including a statement of research and teaching, and a list of at least three references to EMISsearch@smu.edu. Applicants should also make arrangements for their recommendation letters to be emailed directly to EMISsearch@smu.edu no later than January 15, 2017. Nominations of outstanding candidates for Associate and Full Professor positions can be submitted to Dr. Halit Uster, Professor and Faculty Search Committee Chair, uster@smu.edu. SMU is committed to achieving excellence through diversity. The university actively encourages applications and/or nominations of women, persons of color, veterans and persons with disabilities. The committee will notify applicants of its employment decision after the position is filled. Hiring is contingent upon the satisfactory completion of a background check. SMU will not discriminate in any program or activity on the basis of race, color, religion, national origin, sex, age, disability, genetic information, veteran status, sexual orientation, or gender identity and expression. The Executive Director for Access and Equity/Title IX Coordinator is designated to handle inquiries regarding nondiscrimination policies and may be reached at the Perkins Administration Building, Room 204, 6425 Boaz Lane, Dallas, TX 75205, 214-768-3601, accessequity@smu.edu.

ormstoday.informs.org

SPECIAL ADVERTISING SECTION | View Classifieds Online at: http://www.orms-today.org

CLASSIFIEDS

Faculty Positions in the Department of Industrial Engineering and Operations Research Columbia Engineering invites applications for multiple faculty positions in the Department of Industrial Engineering and Operations Research. Applications are sought from candidates with research interests in all areas of operations research and related disciplines. All ranks will be considered. Candidates with research interests in Applied Probability, Data/Business Analytics, and Financial Engineering are especially encouraged to apply. Successful candidates are expected to build a strong methodological research record and to contribute to the Department’s educational programs. Columbia encourages multi-disciplinary research and collaborations across academic units on the campus. The Department is particularly interested in qualified candidates who can contribute to the diversity and excellence of the university community. For additional information and to apply, please see: http://engineering.columbia.edu/faculty-job-opportunities. Applications should be submitted electronically and include the following: curriculum-vitae including a list of publications, a description of research accomplishments, a statement of research and teaching interests and plans, contact information for three experts who can provide letters of recommendation, and up to three pre/reprints of scholarly work. All applications received by December 15, 2016 will receive full consideration. Junior candidates presenting at the INFORMS Annual Meeting are encouraged to submit their application by November 10. Applicants can consult www.ieor.columbia.edu for more information about the Department. Columbia University is an Equal Opportunity/Affirmative Action employer–Race/Gender/Disability/Veteran.

CALL FOR NOMINATIONS AND APPLICATIONS 2017-2018 IEMS Visiting Professor of Instruction

The Department of Industrial Engineering and Management Sciences (IEMS) at Northwestern University invites applications for the Teacher in Residence Program. Teachers in Residence will be experienced faculty who wish to spend a year with IEMS teaching undergraduate courses and interacting with top-quality researchers in the department. We expect that the exchange of ideas and best practices will benefit both IEMS and the Teacher in Residence. Northwestern IEMS boasts a top-five PhD program, as well as a top-ten undergraduate program, and its faculty are respected world-wide for their leadership in a number of research areas. Teachers in Residence will be invited to collaborate in research with current faculty while supporting the undergraduate program by teaching foundational courses and advising senior design projects in the program. Opportunities to teach special topics courses or to develop new courses that can serve both the Northwestern population and the Resident’s home institution are possible. The Department is also interested in co-curricular activities (such as undergraduate research, workshops, seminars, competitions, or similar events) that offer the opportunity for Northwestern students and students from the home institution to collaborate. The successful candidate for the Teacher in Residence program should generally be a tenured faculty member on sabbatical from his or her home institution for the 2017-2018 academic year. They will have demonstrated excellence in teaching and an interest in the assessment and improvement of student learning. Teacher in Residence stipends will be commensurate with experience, and the teaching load will be negotiated based on the ideas and opportunities proposed by the candidate. Teachers in Residence will also be provided a discretionary account that can be used for travel, support of undergraduate research assistants, and other professional expenses. Applications should be submitted electronically to https://facultysearch.mccormick.northwestern.edu/apply/index/OTg and should include: • Cover Letter & CV • Statement of purpose, including description of courses the candidate is particularly interested in teaching or developing, as well as proposal for any co-curricular opportunities the candidate wishes to pursue • Statement of teaching philosophy

advertiser index Page # ADVERTISER

E-mail & Web Page

AIMMS info@aimms.com www.aimms.com

9, 11, INFORMS 14, 27, informs@informs.org 35, 43, meetings@informs.org 44, 57 www.informs.org

17 FICO FICOAmericas@fico.com www.fico.com/shell

3, Frontline Systems, Inc. 5, info@solver.com 7 www.solver.com

LINDO Systems, Inc.

info@lindo.com www.lindo.com

Lumina Decision Systems, Inc.

info@lumina.com www.lumina.com

C4 GAMS Development Corp.

sales@gams.com www.gams.com

• Teaching evaluations for at least the last four years • List of references familiar with the candidate’s experience and interest in teaching and learning. Questions may be directed to tir@iems.northwestern.edu Screening of applications will begin on December 15, 2016. Candidates should submit application materials before this date to ensure full consideration, but applications will be accepted until the position is filled.

Gurobi Optimization

713.87.9341 info@gurobi.com www.gurobi.com

Northwestern University is an Equal Opportunity, Affirmative Action Employer of all protected classes including veterans and individuals with disabilities. Women and minorities are encouraged to apply. Hiring is contingent upon eligibility to work in the United States.

December 2016

ORMS Today

| 63

ORacle

Doug Samuelson

samuelsondoug@yahoo.com

The parable of the termites The group of OR/MS analysts had reconvened once again, shortly after the Presidential election. “Well, Ben,” Brett said, “it looks as if all the quant experts were wrong about this one, doesn’t it?” “Not all,” Ben responded. “John Zogby, one of the best pollsters in the country, said when the FBI memo came out that Hillary had already been on a downward trend for a week or more before that. And Allan Lichtman’s ‘13 Keys’ model was calling it close, and eventually predicting a Trump win, when the polls had Hillary with a double-digit lead. “But you know what else is interesting? Those two don’t call themselves OR/MS analysts! Zogby is a historian – without a non-honorary doctorate – and Allan Lichtman usually describes himself as a quantitative historian. And the fact is, lots of people, not just OR/MS analysts, relied too heavily on trend analyses combined with polls, so when the polls went wrong, so did everything else.That’s how Nate Silver came up, the morning of Election Day, with a probability of 75 percent of Hillary winning. And other ‘experts’ called it as high as 99 percent.” “So it wasn’t just totally swung by the FBI memos?” Brett asked. Ben replied, “No! The FBI memo was like a guy kicking a wall in an old house, and having the wall immediately fall down. Of course it wouldn’t have fallen down right then if he hadn’t kicked it. But it wouldn’t have fallen down from one kick if it hadn’t been full of termites. The termites are the main problem, not the kick. “And, of course,” Ben added, “you know that Hillary and her people are still blaming the FBI director, the media, dirty tactics by the other side – anything except that they just plain missed where some key voting blocs were going! And this after the Democratic Party chairs in those Upper Midwest states had been screaming for weeks that they needed some new ad buys and personal appearances with a strong 64 | ORMS Today

December 2016

as some people I know have contemplated, I think it’s time to take a hard look at the OR/MS profession. Nowadays I call myself a decision scientist, or just a proven trustworthy problem-solver, and forget about trying to

The fact is, lots of people relied too heavily on trend analyses combined with polls, so when the polls went wrong, so did everything else.

economic message! And meanwhile Trump thinks the outcome validated everything he did, so he and his people aren’t looking at lessons learned, either. In fact, some of them had read the same polls and were surprised when they won, which helps to explain the chaotic transition.” “Wow! Impressive,” Jim grunted sarcastically. “Even worse than that,” Ben grimaced, “is the reaction I’ve gotten from some people in the profession to my own analyses. I pointed out before the election that both campaigns were concentrating on states Hillary had been thought to have locked up – a sure sign that her people suspected she was in trouble there. And the response was, ‘Interesting, but where’s the O.R.?’” “What?” Brett exclaimed, clearly puzzled. “I’ll tell you where the O.R. is,” Ben growled. “All too often, the kind of O.R. that insists on higher math and ignores what works ends up in the windowless rooms in the basements, as Russ Ackoff predicted 35 years ago, while other professions take over the actual business of influencing important decisions. And then those over-specialized analysts grind out articles and presentations lamenting the foolishness of the decision-makers who don’t realize that they should value O.R. They ignore evidence, sticking to models that have stopped working, and then blame everyone else for the bad results.Where else have we seen this lately?” “Ugh,” Jim and Brett chorused. “Sounds like it’s a pandemic. So what do we do about it?” “Well, I don’t know about anyone else,” Ben said softly,“but I learned a long time ago not to keep trying to sell analysis to people who are too stubborn or stupid or doctrinaire to value it. Let the people who are responsible for electing this administration deal with its problems.Whether or not I leave the country,

explain what ‘operations research’ is. O.R. in its heyday was applied problem-solving by interdisciplinary teams. Nobody had an O.R. degree or any preset definition of what methods were useful.The more defined the discipline becomes, the less useful it is – and the more resistant it is to being dragged back to reality.“ “You sound upset,” Jim remarked. “Are you sure about this?” “Yeah,” Ben affirmed. “Now’s the time when good O.R. could be tremendously helpful to the country. If O.R. is a uniquely productive way of looking at the world, then the future of the profession is bright. But if it’s codifying a set of ways to analyze and then ossifying around those, it might be doomed. Fortunately, there have been a few people like Gene Woolsey and Russ Ackoff and Saul Gass to remind us how to do it, but I fear people like them are scarce and getting scarcer.” “That sounds serious,” Jim exclaimed. “I think we’ll survive,” Ben shrugged. “But those exceptional analysts, willing to get immersed in the real problems, tackle them creatively and notice evidence of going wrong – are our hope for the future. Meanwhile, the campaign hotshots, some members of the OR/MS profession, and yes, Donald Trump and his top staff should remember an old proverb:The most important learning you’ll ever do happens shortly after you know it all!” ORMS Doug Samuelson (samuelsondoug@ yahoo.com) is president and chief scientist of InfoLogix, Inc., in Annandale, Va. He is a veteran of a number of political campaigns and related analyses in addition to his long career as a federal policy analyst and consultant.

ormstoday.informs.org

“ Everything that’s wrong with the common spreadsheet is fixed in Analytica.” PC Week “ The program itself is close to perfection. The best decision analysis software yet available.” MacWorld “ Analytica is very easy to learn. Once learned it is delightful to use.” J. Human and Env. Risk Assessment

Why leading analysts use Analytica instead of spreadsheets: Clarity: Create models with intuitive influence diagrams.

Scalability: Cope comfortably with models too complex for spreadsheets.

Speed: Run models 10 times faster _ develop models in a quarter of the time.

Power: Use comprehensive libraries for statistics, forecasting, decision analysis, and optimization.

Flexibility: Manage multiple dimensions with ease using Intelligent Arrays™. Risk: Manage uncertainty and risk with efficient Monte Carlo.

Access: Deploy models instantly to the web with Analytica Cloud.

Introducing Analytica 5.0 Snappy new influence diagrams and interaction, flexible array formatting, fast multi-thread computing, Analytica Cloud Player, and much much more.

Download Analytica Free 101 from www.Lumina.com/ORMS

www.lumina.com | 650-212-1212

OPTIMIZATION GENERAL ALGEBRAIC MODELING SYSTEM High-Level Modeling The General Algebraic Modeling System (GAMS) is a high-level modeling system for mathematical programming problems. GAMS is tailored for complex, large-scale modeling applications, and allows you to build large maintainable models that can be adapted quickly to new situations. Models are fully portable from one computer platform to another.

State-of-the-Art Solvers GAMS incorporates all major commercial and academic state-of-the-art solution technologies for a broad range of problem types.

GAMS Integrated Developer Environment for editing, debugging, solving models, and viewing data.

Effects of Proposed Trade Policies on Employment The Peterson Institute for International Economics

Percentage private sector job loss by state

(PIIE) presented the results of a pre-election analysis of the economic implications of the proposed trade policies of the two presidential candidates: Clinton and Trump. The analysis traces the impacts of major changes in trade policy

investment,

macro

aggregates:

government

consumption,

expenditure,

and

international trade. The results indicate that any shock to US international trade has serious effects on employment, including many sectors indirectly linked to exporting industries – when workers lose their jobs, they no longer go to restaurants.

<3.5%

3.5-3.9 ¯

4.0-4.49 ¯ n

4.5-4.9 ¯ 5.0%>

Most-affected US counties The modeling framework contained several components, including a GAMS input-output model, a Python module to disaggregate the results to the state and county level, and a GIS platform to display the results. For further information please visit: https://goo.gl/2ghNx5

High percentage job loss (full trade war scenario)

High job loss, business services sector (asymmetric trade war scenario)

High absolute job loss (full trade war scenario)

High job loss, soybeans sector (asymmetric trade war scenario)

www.gams.com

High job loss, aerospace sector (asymmetric trade war scenario)

High job loss in specific sector (asymmetric trade war scenario) and either percentage or absolute job loss (full trade war scenario)