Page 1

ANALYTICS AND ETHICS: When good A.I. goes bad

February 2018

Volume 45 • Number 1

Vehicle routing software:

Innovation delivers the goods Also Inside: Future directions at NSF O.R. and music generation Measuring medical policy changes Self-service analytics

What’s Your StORy? Jack Kloeber Principal KROMITE LLC INFORMS member since 1991 What prompted you to enter this field? Why? I entered operations research because it was a natural outlet for my initial work on neural network investigation into admission policy, and developing measures and assessment methodologies for candidate military proposals. I already had an MS in Industrial Engineering, so O.R. was already in my blood. What are you most excited about as chair of the 2018 INFORMS Business Analytics Conference? I am most excited about being part of an event that brings such talented and passionate people together to learn and discuss a field that is exploding in its impact, understanding, and use around the world...and we have a chance of being at the top of this field! The Keynote speakers and the Invited Tracks speakers are such experienced analysts and leaders who have done excellent work and are willing to share it with all of us. What INFORMS member benefit do you find the most useful? Attendance at the conferences is my biggest benefit from INFORMS membership. My personal and professional networks are richer and work better because of these conferences. I especially like the networking built into the Analytics conference every April. I have found new vendors, new hires, new partnerships, and certainly, I have found solutions to some of my problems. In addition, I get to see my old friends. Tell us about your experience as an Edelman Laureate and how it has affected your career. As an owner of a small consulting business, after the Edelman Award we immediately started speaking to new organizations/potential clients at a different level. They seemed more open to our ideas and thoughts. After intently studying and watching the other finalists, our team has a broader idea of how we can help, where we can help, and a belief that we CAN help. We were recently invited to talk to the Gates Foundation about agriculture efficiency – a meeting that happened almost entirely because of our work with Syngenta and the Edelman competition. Tell us something that not many people know about you. TWO things: My specialty in the Army was Target Acquisition for Field Artillery, and, I speak Greek. More questions for Jack? Ask him in the Open Forum on INFORMS Connect!


Your Preferred Medium for doing Optimization Based Planning? What’sBest! is pre-customized for your purposes.

What’sBest! allows you to quickly and easily

build powerful optimization planning models in Excel.What’sBest! handles all types of optimization problems, and adds a variety

of additional capabilities to simplify your planning process. It has special features like:       

Easy representation of piecewise linear functions, e.g., to represent power output of a hydro-electric unit.

“Semi-continuous” variables to represent minimum-buy quantities. “All-different” constraints to choose unique position numbers for various objects. “Positive Definiteness” constraint type to help in estimating valid covariance matrices. Network representation capabilities for modeling multi-level supply chains, Special functions for modeling heat transfer in heat exchangers. Special functions for modeling fluid flow in networks.

Global solver for handling non-convex functions such as in Optimal Power Flow in electricity distribution.  Robust optimization support of convenient functions such as MIN(), MAX(), ABS(), VLOOKUP, SUMIF, etc.  Scenario generation capabilities for stochastic planning under single and multi-stage uncertainty. 

Find out how easy it is to get started.

Call: 312-988-7422 Visit: email:

Contents February 2018 | Volume 45, No. 1 |

20 On the Cover Driving innovation Biennial survey of vehicle routing software focuses on recent and future trends. Image Š alphaspirit |

F e at ure s 20

When good A.I. goes bad


Future directions at NSF


O.R. and music generation


Measuring medical policy changes

By Joseph Byrum Analytics ethics: Artificial intelligence will make mistakes, but validation is the key to avoiding algorithmic disaster.

By Sheldon H. Jacobson, Jerome F. Hajjar, Andrew Johnson, Erick Moreno-Centeno and Siqian Shen Intellectual merit, broader impact require attention if NSF is to continue to contribute to U.S. economic and social well-being.

de partm e nt s

6 Inside Story 8 President’s Desk 10 INFORMS in the News 12 INFORMS Initiatives 14 Newsmakers 16 Issues in Education 18 Viewpoint 59 Classifieds 62 Industry News 64 ORacle

By Dorien Herremans and Elaine Chew The role of operations research in state-of-the-art automatic composition systems: challenges and opportunities.

2 | ORMS Today

By Sumana Reddy and Harrison Schramm An O.R. application to the diagnosis and treatment of depression: insights for small- to mid-sized medical practices. |

February 2018


Solver – Re-Imagined for the Future Cloud-First Optimization Made Easy for Touchscreens and Mobile

The Most Popular Optimization Tool in History is Ready for the Future – and Available Now, Free. Starting in 1990 with Excel 3.0, now in the hands of over a billion Excel users, Solver has been re-engineered multiple times, while remaining simple to use and compatible across Windows, Mac, and the cloud. Now Solver is new again, easier to use, and easier to upgrade for large-scale models than ever before.

Easy to Use on Touchscreens and Mobile Devices. Solver’s new user interface, now in a “modeless” Task Pane, is designed for both mouse pointers and human fingers on touchscreens and mobile devices.

And when you need to solve larger models, upgrades to the world’s best linear and nonlinear Solvers, handling thousands to millions of variables and constraints, are available immediately – no installation needed.

Use in Excel Online and Excel for Windows or Mac. Solver works equally well in Excel Online (through a browser), and Excel 2016 for Windows or Mac. And your models are compatible across all Excel versions.

Nothing to Install – Just Insert in Your Workbook.

Solves Every Type and Size of Optimization Model.

It’s super-easy: Just choose Insert and under Add-ins, click Store and search for Solver. Visit to learn more, ask questions, or start a free trial of our most advanced tools for Excel, Tableau and Power BI.

Solver diagnoses your model type – linear, quadratic, nonlinear, non-smooth – and helps you select the right Solver Engine. Small-scale linear and mixed-integer programming, nonlinear and global optimization, and our Evolutionary Solver are built-in (and free).

Tel 775 831 0300 • Fax 775 831 0314 •

February 2018 | Volume 45, No. 1 |


INFORMS Board of Directors

President Nicholas Hall, Ohio State University

President-Elect Ramayya Krishnan, Carnegie Mellon University

Past President Brian Denton, University of Michigan

Secretary Victoria Chen, University of Texas at Arlington

Treasurer Michael Fu, University of Maryland

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

F e at ure s 38

Self-service analytics By Paul Brunet As SSA gains momentum, the need for data governance increases in order to drive true business value.

So ft ware Sur v e y 42

Vice President- Russell Barton, Sections and Societies Pennsylvania State University

Vice President- Marco Luebbecke, Information Technology RWTH Aachen University

Vice President- C. Allen Butler, Practice Activities Daniel H. Wagner Associates

Vice President- Sue Merchant, Blue Link Consulting International Activities

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

Vehicle routing

Editors of Other INFORMS Publications

By Peter Horner Side-by-side comparison of packages: Vehicle routing software developers are always moving forward … or they’re run over.



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

Editor’s Cut Anne G. Robinson, Verizon

Information Systems Research Alok Gupta, University of Minnesota INFORMS Journal on Computing David Woodruff, University of California, Davis I NFORMS Journal on Optimization Dimitris Bertsimas, Massachusetts Institute of Technology

INFORMS Transactions Jeroen Belien, KU Leuven on Education

Interfaces Michael F. Gorman, University of Dayton Management Science David Simchi-Levi, Massachusetts Institute of Technology


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 John R. Birge, University of Chicago

Organization Science Gautam Ahuja, Cornell University

Service Science Paul P. Maglio, University of California, Merced Strategy Science Daniel A. Levinthal, University of Pennsylvania Transportation Science Martin Savelsbergh, Georgia Institute of Technology

47 Analytics Conference preview

50 Roundtable Report

51 Subdivision Awards

47 Continuing Education workshop

57 Simulation Society awards

49 International Conference preview

58 People

58 Meetings

4 | ORMS Today


February 2018

Tutorials in Operations Douglas Shier, Clemson University Research

INFORMS Office • Phone: 1-800-4INFORMS

Executive Director Melissa Moore


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

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

Everything in Predictive and Prescriptive Analytics Everywhere You Want, from Concept to Deployment.

functions; easy multiple parameterized simulations, decision trees, and a wide array of charts and graphs.

The Analytic Solver® suite makes the world’s best optimization software and 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.

Forecasting, Data Mining and Text Mining.

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.

Comprehensive Risk and Decision Analysis. 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

Analytic Solver is also a full-power, point-and-click tool for predictive analytics, from time series methods to classification and regression trees, neural networks, and access to SQL databases and Spark Big Data clusters.

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 to learn more or ask questions, and visit to register and start a free trial – in the cloud, on your desktop, or both!

Tel 775 831 0300 • Fax 775 831 0314 •

Inside Story

Peter Horner, editor

Ethics, O.R. & analytics Anyone with a conscience believes in ethics – in their personal life, in their business life, in their any other life. The definition of ethics varies depending on the dictionary, but basically it boils down to doing the right thing, even if that “thing” might cost you your job, money, prestige, friendships, whatever you may hold dear. When its Cer tified Analytics Professional (CAP®) program was launched several years ago, INFORMS included an ethics component that outlined norms of behavior that CAP recipients were expected to uphold. In 2016, INFORMS issued a set of ethics guidelines for those working in the O.R. and analytics field to “aspire to.” INFORMS members David Hunt and Scott Nestler spearheaded the creation of an INFORMS Ethics in O.R. & Analytics Group to promote the guidelines. “ Wo r k d o n e by m e m b e r s o f INFORMS increasingly impacts peoples’ lives as the use of our algorithms and models spreads and as data collection becomes more detailed and more personal,” Hunt and Nestler wrote in an article for OR/MS Today. “Yet too often we become absorbed by the elegance of our models and research, failing to consider any potential ethical implications. Or, we find ourselves faced with a situation that seems unethical, and are forced to determine the best course of action without previously having established exactly what we consider to be ethical.” I n t h i s m o n t h ’s l e a d f e a t u re (“When good A.I. goes bad,” page 20), contr ibuting author Joseph Byrum brings up ethics as it relates to artificial intelligence and machine learning, two of the hottest topics in the O.R. and analytics orbit right now. Writes Byrum: “The lesson as we develop increasingly powerful A.I. solutions is that the A.I. can never substitute for due diligence. The 6 | ORMS Today


February 2018

more we turn over decisions to A.I., the more we must ensure that procedures are in place to catch the mistakes made not by the A.I., but the humans who set up and operate the systems.” In other words, Byrum says, validation and verification remain crucial. The feature lineup is backed by our biennial survey of vehicle routing software (page 42). Given the widespread media coverage of driverless vehicles and delivery drones, I thought it would be insightful to ask survey respondents what impact, if any, drones and driverless vehicles would have on VR software in the future. Here are a few of their responses: • “Planning will probably not change, but more intelligent systems will need to be created to interact with them for execution.” - Christian Lafrance, Clear Destination • “If drones and driverless vehicles deliveries do become mass market, the routing problem is essentially the same but will likely require new constraints and customizations.” - Phil Welch, Open Door Logistics • “Drones and driverless vehicles will add new constraints to existing systems, but their impact will be minimal. Other future changes will have much bigger impact.” - Felipe Carvalho,Widescope • “Driverless technology will be subject to the same constraints in the transport network as a human driver. Good savings opportunities come with shifting solutions in the time domain.The main advantage is that drive-time constraints and other human constraint factors can be improved. For driverless vehicles, existing algorithms would be able to find new optimizations that may be infeasible at the moment.” - Ruben Filter, Intelligent Routing – ORMS

OR/MS Today Advertising and Editorial Office Send all advertising submissions for OR/MS Today to: Lionheart Publishing Inc. 1635 Old​41 Hwy, Suite 112-361, Kennesaw, GA 30152​USA Tel.: 888.303.5639 • Fax: 770.432.6969

President & Advertising Sales John Llewellyn Tel.: 404.918.3275

Editor Peter R. Horner Tel.: 770.587.3172

Assistant Editor Donna Brooks

Contributing writers/editors Douglas Samuelson

Art Director Alan Brubaker, ext. 218

Online Projects & Reprints Manager Patton McGinley, ext. 214

OR/MS Today Committee James Cochran, chairman


Lionheart Publishing Online

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, 1635 Old​ 41 Hwy, Suite 112-361, Kennesaw, GA 30152​. 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 ©2018 by the Institute for Operations Research and the Management Sciences. All rights reserved.

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 7.5 State-of-the-Art Mathematical Programming Solver Experience state-of-the-art optimization algorithms backed by industry-leading customer support. +1 713-871-9341

Contact us today to request a FREE full-featured, unrestricted evaluation license.

President’s Desk

Nicholas G. Hall

INFORMS President

INFORMS goes to Washington: Policy Days and advocacy

Ever ything INFORMS does is motivated and advised by our four strategic goals: • INFORMS will identify, recognize and promote the work of our members to show the value their science and practice brings to society. • Decision-makers will have access to, and use, innovative technologies and methodologies to transform visions, tasks or responsibilities into better choices, services and products to achieve better outcomes. • Organizations will identify operations research and analytics as core components of success and institutionalize operations research and analytics input as part of their decision-making processes. • Operations research and analytics will advance society and make the world a better place. Within the context provided by these goals, INFORMS now finds itself in a uniquely positive situation. Whether measured by our best-ever financial position, close to record membership numbers or remarkable research impact, INFORMS has never been stronger.This is due in large part to the outstanding leader ship of my predecessor s as president and as Board members, as well as our headquarters staff. As a result, INFORMS is now able to move beyond day-to-day constraints and issues, and focus on larger, more strategic initiatives. For 2018, these initiatives are visibility and awareness. 8 | ORMS Today


February 2018

We are identifying topics that are of greatest importance to policymakers and where O.R. and analytics can provide value.

Visibility Visibility refers to a three-year (2018-2020) program of live “Policy Day” events to be held in Washington, D.C., for policymakers and senior administrative staffs. The 2018 Policy Day event is being managed by a committee chaired by Laura Albert, and includes Jeff Cohen, Brian Denton, Ramayya Krishnan, Melissa Moore and myself. In planning this event and identifying the right topics, speakers, audience and venue, we are fortunate to have the guidance of Signal Group, a leading consulting and lobbying firm in Washington. The objectives of the Policy Day event are as follows: • Increase awareness of how O.R. and analytics can be used in policymaking and solving major societal problems. • Increase awareness of how O.R. and analytics can be used within the federal government. • Attract well-positioned congressional and agency staff. • Attract domain-relevant directors and advisory board members from government agencies (e.g., NSF, NIH, DOE, DOE). • Increase requests for INFORMS experts to assist in problem-solving. • Involve INFORMS members, for example potentially live streaming the Policy Day, archiving presentations on YouTube, developing a website on INFORMS Connect and publishing an OR/MS Today article after the event.

• Increase engagement with INFORMS members who are senior officials at government agencies. We are identifying topics that are of greatest importance to policymakers and where O.R. and analytics can provide value in the form of improved solutions, as well as examining various possible formats for the Policy Day event. The overall message we will leave behind at the end of the Policy Day is: “Operations research and analytics have unique and effective impact and value in solving problems of national and international importance.” We are optimistic about the impact of this event, and we will also learn from it to improve those that will follow. Awareness Awareness refers to an ongoing program of advocacy to policymakers about the relevance and expertise of O.R. and analytics, INFORMS and our members. Awareness is a multiple-year initiative with a strong, self-learning process, and again with the guidance of Signal Group. INFORMS’ strategic advocacy plan consists of three integrated elements: Congressional advocacy: The purpose of congressional advocacy is to build

The advocacy plan will include meetings on Capitol Hill, with members outside of D.C. and with local press to promote O.R. awareness, interest, relationships and support for the value proposition of O.R. and analytics among policymakers. The advocacy plan will include meetings on Capitol Hill and with members outside of D.C., and with local press to promote O.R. and analytics work within specific congressional districts or states. We will initially focus on the House Committee on Science, Space and Technology and the Senate Committee on Commerce, Science and Transportation, which make authorizations to NSF. We also intend to utilize other congressional organizations

that we believe will be interested in our message. For example, the “Problem Solvers’ Caucus” within the House of Representatives i s a b i p a r t i s a n g ro u p of 40 lawmakers, about equally divided b e t we e n R e p u bl i c a n s and Democrats, who are looking for pragmatic non-partisan ways to help Washington work better – a seemingly perfect place to talk about O.R. and analytics in public policy! Administration advocacy: Beyond Capitol Hill, spreading the word to federal agencies about the impact of operations research and analytics is an important component of building allies and advocates for INFORMS. We will work with Signal Group to determine a path for engaging with senior career officials in targeted agencies. Also, we

will leverage the combined expertise of our members who have direct experience with various agencies. Indirect advocacy: We will develop an outreach strategy that focuses specifically on federal public policy media. This will include developing and placing story ideas and op-eds, and conducting in-person briefings with key editors. INFORMS will develop a high-level brochure that will have effective skim-and-pass-along value among policy peers, along with a white paper that offers more detailed but nontechnical information about the impact and value of O.R. and analytics. We welcome the involvement and support of members in these ambitious initiatives. If you are interested in helping us with information, content development, introductions, comments or suggestions, please contact me at I look forward to hearing from you. ORMS

EARLY CAREER PROFESSIONALS’ NETWORK (ECPN) Nominations Deadline: February 28, 2018

• ECPN is the former INFORMS Professional Colloquium (IPC) and Early Career Connections (ECC) combined. • Focused on early-career professionals across all types of organizations (e.g. industry & academia) • Designed to help workshop participants transition into successful careers and network with their cohort as well as with more experienced O.R. leaders and practitioners from diverse backgrounds

For More Information: Visit: Email: February 2018


ORMS Today

| 9

INFORMS in the News

Compiled by Ashley Kilgore

Edelman finalists, Super Bowl ads, politics and more Making government work: New office strengthens role of data at FCC The Federal Communications Commission (FCC), which pioneered the use of auctions to allocate spectrum for wireless service, has been named one of six finalists for the prestigious INFORMS Franz Edelman Award for Achievement in Operations Research and the Management Sciences.  - Forbes, Jan. 10

Europcar selected as 2018 Edelman Award finalist Europcar is one of six finalists for the 47th annual Franz Edelman Award for Achievement in Operations Research and the Management Sciences from INFORMS, an operations research and analytics association. - Auto Rental News, Jan. 9

FCC, TBS cited for analytics prowess The FCC and Turner Broadcasting are among those listed among the six finalists for the prestigious 2018 INFORMS Franz Edelman Award for Achievement in Operations Research and the Management Sciences. - Multichannel News, Jan. 5

Are Super Bowl ads worth it? New research shows that the value of Super Bowl ads can persist beyond the

Image © Think Stock

10 | ORMS Today


February 2018

big game into the basketball and baseball seasons. The 2018 Super Bowl was held Feb. 4, and more than 40 percent of U.S. households watched the game on TV. With a 30-second TV commercial priced at over $4 million, some advertisers may debate whether these ads are worth the cost. A new study in the INFORMS journal Marketing Science finds that the benefits from the Super Bowl ads persist well into the year with increased sales during sporting events such as NCAA’s “March Madness,” the NBA playoffs and MLB games. - INFORMS, Jan. 5

INFORMS announces 2018 Franz Edelman Award finalists INFORMS, the leading international association for professionals in operations research and analytics, has selected six finalists for the 47th annual Franz Edelman Award for Achievement in Operations Research and the Management Sciences, the world’s most prestigious award for achievement in the practice of analytics and O.R.  - Business Insider, Jan. 4

Researcher solves college football’s biggest mystery: recruiting There is an entire industry built up around deciphering where 16- and 17-year-olds will play college football. Websites boast “crystal ball” predictions of where top high school recruits will suit up. Companies charge for premium subscr iptions with claims that they can decode the caprice and whimsy of children. However, a new mathematical model can predict with 70 percent accuracy where a high school football player will go to college using nothing but their basic biographical information and Twitter account. The paper on these

findings was published in the INFORMS journal Decision Analysis. - The Washington Post, Dec. 20. 2017

Online ad ban has limited impact on consumer access to foreign pharmacies As American consumers turn to online search eng ines for cheaper p re s c r i p t i o n d r u g s f ro m f o re i g n pharmacies, safety and quality concerns arise. In 2010, under compulsion from the Department of Justice, Google agreed to forfeit $500 million and ban sponsored search advertising by pharmacies that are not certified by the National Association of Boards of Pharmacy (NABP). The ban covered all foreign phar macies and a few domestics. Within the year, other major search engines followed. But a forthcoming study in the INFORMS journal,  Marketing Science, a leading scholarly marketing publication, finds that the ad ban had limited success in reducing access; motivated consumers still gained access to foreign pharmacies through organic search links.  - INFORMS, Dec. 15, 2017

Music streaming sites benefit indie singers at expense of top 100 artists While free or low-cost music streaming sources like Spotify decrease the use of paid music platforms such as iTunes, a new study in the INFORMS journal  Marketing Science shows they significantly increase exposure for and access to lesser known or indie artists that fall outside the top 100 or even top 500 listings. - INFORMS, Dec. 14, 2017

All politics – and cannabis marketing – are local California’s legal cannabis market, opening for business on Jan. 1, is expected to quickly grow to be the largest in the nation and worth more than $5 billion a year. County voting on Proposition 64 that led the state here – to legalizing sales for recreational use – can offer insight into how medical marijuana dispensaries will now market themselves, according

to a new study in the INFORMS journal Organization Science. - UC Davis News, Dec. 14, 2017

New tool to improve kidney transplant success, reduce costs Waitlists for life-saving kidney transplants are long and every donated kidney is precious. That’s why patients on the waitlist are screened regularly to ensure their suitability for transplant. But is the screening process as effective as it could be? A new study in the INFORMS journal Operations Research, conducted by INFORMS members Steven Shechter and Woonghee Tim Huh, both of UBC Sauder, and Alireza Sabouri of the University of Calgary, devised a strategy – the first evidence-based tool of its kind – that would reduce the number of risky transplants by 23 percent and reduce the annual number of screenings by 27 percent. - UBC News, Dec. 6, 2017

data on New York City’s bike sharing system, CitiBike, at the INFORMS Annual Meeting in Houston in October 2017. - Penn State News, Dec. 4, 2017

Why stand in line on Black Friday? The psychology explained I N F O R M S Fe l l ow a n d M I T professor Richard Larson provides interesting insight into the unique draw of the long lines and even longer waits that shoppers are willing to endure in order to score a Black Friday deal. - The New York Times, Nov. 23, 2017

Hidden cost of crime: Tanzanians pay to protect money from theft “Mobile money,” a checking account attached to a mobile phone number, has revolutionized the financial lives of millions of people in many developing countries without access to a banking infrastructure. Using text messages or apps on consumer phones and in partnership with mom-and-pop retailers, who serve as cash-in and cash-out points, telecom companies have created an accessible and cost-effective virtual banking infrastructure in many developing countries, according to a new study in the INFORMS journal Marketing Science. - INFORMS, Nov. 13, 2017

Image © Inge Hogenbijl |

Suffering artists work sells for 35% less than their happier works According to a new study in the INFORMS journal Management Science, personal unhappiness, particularly that exper ienced in times of mour ning or bereavement, can actually cause a significant decrease in the value of an artist’s work. - Daily Mail, Dec. 4, 2017

Engineering management student presents research at national conference An international student in the Master of Engineering Management program at Pennsylvania State University, Samarth Patel, had a unique opportunity to present

Why supply chain professionals still need ‘soft skills’ in the technology era “Soft skills” such as communication and teamwork are crucial proficiencies for a supply chain leader to hold. However, these skills are often overlooked amid today’s technology-dr iven industry, according to Yossi Sheffi, Ph.D., INFORMS member and professor of engineering systems at the Massachusetts Institute of Technology in Cambridge, Mass., and director of the MIT Center for Transportation and Logistics. - Becker’s Hospital Review, Nov. 9, 2017

What’s the cause of stock market crashes? Too much testosterone, science says The financial institutions that do work trading on the New York Stock

Image © Setsiri Silapasuwanchai |

Exchange are predominantly, perennially staffed by men. And now, a new study from the University of Western Ontario, University of Oxford and Claremont Graduate University, published in the INFORMS journal Management Science, has found that testosterone, the hormone found more so in men than in women, can be linked to the decisions that lead to stock market destabilization and crashes. - Entrepreneur, Nov. 11, 2017

Cheat sheet: How to become a data scientist Data scientists are in high demand, taking the coveted No. 1 spot on Glassdoor’s Best Jobs in America list in both 2016 and 2017. INFORMS is highlighted as an organization where those interested in exploring the field of data science can seek out information about the profession as well as certification and training solutions. - TechRepublic, Nov. 7, 2017

Organization appointments: Ramayya Krishnan R a m ay ya K r i s h n a n , d e a n o f the Heinz College of Infor mation Systems and Public Policy at Carnegie Mellon University, has been elected as president of the Board of Directors at INFORMS. ORMS - The Chronicle of Higher Education,  Nov. 5, 2017 Ashley Kilgore is the public relations manager at INFORMS.

REFERENCES The complete list of references cited in this article are available online at:

February 2018


ORMS Today

| 11

INFORMS Initiatives

Edelman finalists, Pro Bono Analytics, UPS Prize finalists Six finalists named for Edelman Award INFORMS selected a diverse group of six finalists for the 47th annual Franz Edelman Award for Achievement in Operations Research and the Management Sciences, the world’s most prestigious award for achievement in the practice of analytics and O.R. The 2018 finalists, who will present their work before a panel of judges at the INFORMS Conference on Business Analytics & Operations Research in Baltimore on April 15-17, included innovative applications in broadcasting, healthcare, communication, inventory management, vehicle fleet management and alternative energy. Fo r m o re t h a n f o u r d e c a d e s , winners of the Edelman Award have been recognized for transforming how to approach some of the world’s most complex problems. Finalists for the

Edelman Award have contributed a cumulative impact of more than $250 billion since the award’s inception. The finalists for the 2018 Edelman Award are: China National Petroleum Corporation (CNPC): China’s natural gas consumption has nearly doubled over the past five years. To better meet demand, CNPC, China’s largest oil and natural gas producer and supplier, partnered with researchers from the University of California at Berkeley and Tsinghua University to develop and implement a new software that optimized the country’s natural gas pipeline. Previously, all annual planning for CNPC, which controls 75 percent of the country’s natural gas resources and pipeline network, was manually conducted using spreadsheets. However, the increasing complexity and size of China’s natural gas pipeline meant that the previ-

The Franz Edelman Award is the world’s most prestigious award for achievement in the practice of analytics and O.R.

12 | ORMS Today


February 2018

ous method had large measures of error, increasing costs and wasting resources. Since implementing its new systems at the end of 2014, CNPC has realized approximately $330 million in direct savings for CNPC, and the increased efficiency of the pipeline has enabled it to delay further pipeline expansions saving billions of additional dollars. Europcar: Europcar, the leading European car rental company that provides rental services to more than 5 million drivers annually, partnered with ACT Operations Research to create Opticar – an advanced logistics management solutions for tackling the many factors that impact the car rental industry each day, such as the need for vehicles at airports and hotels and customer requests for vehicle upgrades. The system can also anticipate future demand for Europcar’s fleet of vehicles up to six months in advance, improving capacity management. In addition, the Opticar system enabled Europcar to optimize its approach to revenue management, providing a new, more stable pricing system for its vehicle rentals that takes into account competitor pr icing, what vehicles are currently available, and expected demand for vehicles. Federal Communications C o m m i s s i o n : The Federal Communications Commission recently completed the world’s fir st twosided auction of valuable low-band electromagnetic spectrum, reclaiming channels from TV broadcasters to meet the exploding demand for wireless services. The commission purchased spectrum from TV broadcasters, sold the acquired spectrum to wireless providers, and assigned the remaining broadcasters to new channels in a smaller TV band. Operations research tools, including optimization software and satisfiability solvers, were essential to the spectrum clearing target calculations, auction winner determinations and final TV channel assignments. The auction repurposed 84 MHz of TV spectrum for

wireless use, raised nearly $20 billion in revenue, paid more than $10 billion to winning broadcasters, provided nearly $2 billion for relocation costs for nonwinning broadcasters, and contributed more than $7 billion to reduce the federal deficit. The final channel assignments, which included all 2,900 U.S. and Canadian TV stations, enabled 78 percent of the stations to remain on their existing channels, providing an estimated savings of more than $200 million in relocation costs. Intel: Intel, which employs more than 100,000 people in more than 70 countries around the world and has an annual revenue of $60 billion, implemented a fully automated multi-echelon inventory optimization (MEIO)-based inventory target-setting system managing $1 billion daily in finished goods inventory representing more than $40 billion a year in sales. Algorithm-derived inventory targets at Intel are accepted by planners more than 99 percent of the time and have simultaneously dr iven higher customer service and lower inventory levels resulting in more than $1.3 billion in gross profit since 2014. In addition, customers are delighted; since MEIO was implemented at all of Intel’s vendor-managed inventory hubs in 2012, customer satisfaction has never been higher and Intel has landed in the top 10 of Gartner’s Supply Chain Top 25 every year. Faculty in the department of Business Analytics and Statistics at the University of Tennessee-Knoxville and the supply chain software company Logility also contributed to this project. Pe d i a t r i c H e a r t N e t wo r k : The Pediatric Heart Network enlisted researchers with the Georgia Institute of Technology to create clinical practice guidelines (CPG) for pre-, intraand post-surgical care of patients with congenital heart defects (CHDs), the most common birth defect, impacting nearly 1 million children and 1.4 million adults in the United States. Substantial variances in surgical practices to treat patients with CHDs among dif-

ferent healthcare centers were reflected in inconsistent surgical outcomes, some of which resulted in negative consequences for patients. By studying the nine leading U.S. pediatric centers, the researchers identified seven significant factors for influencing surgical outcome, and implemented a CPG that enables patients to be removed from breathing apparatuses earlier, lowered the rate of reintubation, and decreased the time patients need to remain in the intensive care unit.These guidelines also realized a cost savings of 27 percent, which translates to $13,500 per patient. Turner Broadcasting System, Inc.: Tur ner Broadcasting System, Inc., a division of TimeWarner that manages multiple popular television networks including CNN, TBS and the Cartoon Network, developed two television audience targeting solutions that simultaneously created advertiser and sales efficiencies. TargetingNOW and AudienceNOW, released in 2014 and 2015, respectively, revolutionized media industr y standards and processes by implementing integrated forecasting and optimization models to increase the demographics and reach of advertising opportunities. Currently, more than 130 advertisement deals have been executed under these new solutions representing significant ad revenue. Turner anticipates that by 2020, half its advertising inventory will be purchased through TargetingNOW and AudienceNOW, realizing ad revenue in the billions. Based on the current success of these systems, Turner is expanding its use of these methods to optimize other areas of media revenue management, including its programing schedules, as well as its Latin American markets. First awarded in 1972, the prize is named in honor of Franz Edelman, who founded the operations research division within RCA, one of the first corporations to embed operations research as a business imperative.

Pro Bono Analytics: Looking back on 2017 By David Hunt In 2017, Pro Bono Analytics moved from an initiative to an official INFORMS program, and the response has been overwhelming. In addition to helping nonprofit organizations across the country and hosting a special fundraiser to support the city of Houston, the hard work of our volunteers is helping to build a quality program of which all INFORMS members can be proud. Pro Bono Analytics continues to focus on its mission of using analytics to make a difference to nonprofit organizations working in underserved areas and for underserved populations. 2017 by the numbers: • seven projects completed • 10 projects underway, carried into 2018 • more than $7,500 raised to benefit Houston Annual Meeting partner organization, plus creation of 1,200 toiletry kits distributed to Houston’s homeless population The Pro Bono Analytics network has grown to more than 500 volunteers, based around the world, making all of this charitable work in their communities possible. Volunteers were able to successfully employ their skills in decision analysis, marketing science, logistics and general analytics to complete projects that helped a wide variety of nonprofit organizations better fulfill their mission. A quick look at a few nonprofit clients: Goodwill Industries, Grand Rapids, Mich.: Volunteers worked with the VP of Business Intelligence to evaluate and make recommendations about various ad campaigns and their impact on instore foot traffic. “We are so excited about the information that was presented to us. We are following the recommendations and trying to schedule the ad coupons in a way that is more measurable. The information on how to structure the data will be so helpful.” February 2018


ORMS Today

| 13

INFORMS Initiatives

UB engineering students whose “Pro Bono Analytics” project delivered results to two local, nonprofit organizations. Photo credit: Meredith Forrest Kulwicki

Meals on Wheels for Western NewYork: Student volunteers from the University at Buffalo focused on optimization for the “pack-out” process for the delivery of hot and cold meals. Their recommended processes will allow for more accurate tracking, enabling staff and volunteers to serve clients more effectively. “All of the students were very, very professional and took a genuine interest in the project.”

Family Services of Greater Boston: Volunteers helped develop data management strategies to standardize collection and coordinate care across programs to better serve clients in a community-based behavioral health service organization. “[Pro Bono Analytics] offered an excellent and professional approach to our statistical and data challenges.”

Fundraising and volunteering in Houston: The 2017 INFORMS Annual Meeting took place in Houston less than two months after the city was devastated by Hurricane Harvey. Pro Bono Analytics partnered with local homeless shelter, Star of Hope Mission, and thanks to the generosity of INFORMS members, conference attendees and corporate donors, more than $7,500 was raised in support of Houston’s

homeless population. During the Annual Meeting, members and attendees volunteered their time to help build more than 1,200 personal hygiene kits that were distributed to homeless men, women and children on the streets of Houston by Star of Hope Mission. Thanks goes out to everyone who joined this mission to make a difference for those who truly needed a helping hand. What’s next for Pro Bono Analytics?

Pro Bono Analytics will continue to take on new projects that provide analytics support to nonprofit organizations, thus helping to make a difference in underserved communities in 2018. Also, after the success of the fundraising and volunteer effort in Houston, there is an initiative to partner with an organization in Phoenix, the 2018 INFORMS Annual Meeting host city, and all future host cities. Want to help on this and other projects? One way is to visit our website at www. and sign up to become a volunteer helping change the world through analytics. New projects are sent to


Management and Business Review: new magazine prepares for launch Management and Business Review (MBR), a new executive magazine co-sponsored by 10 leading business schools, is scheduled to release its initial call for papers in February. The goal of MBR is to bridge management practice, education and research, and thereby enhance all three. The 10 co-sponsoring business schools include: Carnegie Mellon, China Europe International Business School, City University of Hong Kong, Cornell, Indian School of Business, INSEAD, Michigan, UCLA, Vanderbilt and Virginia. MBR’s founding editors-in-chief, Wallace Hopp and Kalyan Singhal, have long ties to INFORMS. Hopp, a business professor and associate dean of the Ross Learning Design Initiative with the Ross School of Business at University of Michigan, is a former editorin-chief of the flagship INFORMS journal Management Science. Singhal, a professor of innovation, operations and supply chain management with the Merrick School of Business at the University of Baltimore, is the editor-inchief of Production and Operations Management. W h e re a s th e H a r va r d B u s i n e s s R evi e w, S l o a n Management Review and California Management Review are published by individual schools, MBR is a grassroots initiative supported by 10 leading business schools and a large team of distinguished advisors, including 10

14 | ORMS Today


February 2018

professors from Wharton, eight from Harvard, six from Dartmouth, five from Carnegie Mellon, four each from MIT and Stanford, three each from Chicago, Columbia, Cornell, Duke, London Business School, Michigan, Northwestern and UCLA and two each from Georgia State, Indian School of Business, NYU and UC-Berkeley;. Advisors also include 11 executives who have published best-selling books and articles in magazines such as the Harvard Business Review. Details of MBR’s plan are posted on its website www. The plan describes how each school can publish its own edition of MBR. The editors-in-chief encourage academicians to share the plan with their dean and colleagues. They also ask interested parties to send, via email to, the name of your school and dean, along with the dean’s email address so that they can keep her/him up to date on the progress of the magazine. As noted, a call for papers will be posted on MBR’s website in February. Feel free to nominate your colleagues or yourself to the editorial team by sending an email to You can also register yourself on the MBR website so that the editors-in-chief can send additional information from time to time. Suggestions can be sent to Kalyan Singhal ( ORMS

all volunteers as they become available – simply submit your resume to any project that piques your interest or matches your skill set. An important feature of Pro Bono Analytics is that the nonprofit organization selects their partner volunteer(s). Another way to help is spreading the word of Pro Bono Analytics. If you or a colleague knows of a nonprofit organization that could benefit from analytics services, feel free to get in touch with the Pro Bono Analytics program manager. INFORMS Pro Bono Analytics extends its gratitude to all those who were involved with the program in 2017.The program will continue to grow in 2018 while achieving its mission of helping nonprofit organizations make a greater difference in their communities through analytics. None of this would be possible without the dedication of volunteers, nonprofit partner organization and the Pro Bono Analytics Committee. For more information about Pro Bono Analytics or if you have any

feedback on the program, contact Miles Feldman ( – David Hunt chairs the INFORMS Pro Bono Analytics Committee.

INFORMS announces UPS Prize finalists I N F O R M S h a s s e l e c t e d t wo finalists for the 2018 UPS George D. Smith Pr ize, which recognizes excellence in prepar ing students to become practitioners of operations research and analytics. The 2018 finalists: • School of Operations Research and Information Engineering, MEng Programs, Cornell University • Haslam College of Business MSBA, University of Tennessee The winner will be announced on April 16 at the 2018 INFORMS Conference on Business Analytics & Operations Research in Baltimore.

Named in honor of the late UPS chief executive officer – a champion of operations researchers at a leading Fortune 500 corporation – the UPS George D. Smith Prize was created in the spirit of strengthening ties between industry and the schools of higher education that graduate young practitioners of operations research. The prize is awarded to an academic department or program for effective and innovative preparation of students to be good practitioners of operations research or analytics. Past award recipients include the U.S. Air Force Academy; H. John Heinz III College of Information Systems and Public Policy at Carnegie Mellon University; Sauder School of Business, University of British Columbia - Center for Operations Excellence; MIT Leaders for Global Operations (LGO); Naval Postgraduate School; and the Tauber Institute for Global Operations at the University of Michigan. ORMS



Denver: June 20–21 University of Denver

Washington, DC: September 19–20

AMA Washington, DC Area Executive Conference Center


for INFORMS flagship continuing education course taught by instructor Patrick Noonan, PhD, Emory University.

Find out more at February 2018


ORMS Today

| 15

Issues in Education

By Vijay Mehrotra

Get out of your comfort zone In my recent MBA elective course on data mining, I assigned my students a term paper. The assignment required students to clearly describe a problem that they thought would be amenable to machine-learning methods and then discuss how this problem might be solved (what techniques they would suggest using, what data would be required and how that data might be captured, what types of benefits would accrue and to who, what barriers to implementation they might encounter, etc.). This was the first time I had tried this assignment. As with any experiment, I was not sure what kind of results it would produce. The students chose topics ranging from reasonably frivolous (a recommendation app for which beer to choose in a crowded bar) to energetically pragmatic (how to choose a second market for an organic grocery delivery company and how to predict the likelihood of a video advertisement going viral) to ambitiously altruistic (how to proactively issue an alert when an artificial heart is likely to fail). The quality level was equally varied. After learning algorithms for classification, prediction and clustering, several students reported that this assignment forced them to be more creative, to think through more complex problems, and to confront data issues at a very different level. “Definitely got me out of my comfort zone,” one of my students told me. After reading these papers, I still had a nagging suspicion that perhaps all of this effort was largely for naught. Most “real” analytics applications have a great deal of complexity, especially in 16 | ORMS Today


February 2018

Scott Hartley would suggest that it is high time we all get over this kind of bias. Hartley, a venture capitalist and start-up advisor, is the author of a recent book entitled, “The Fuzzy and the Techie: How Liberal Arts Will Rule the Digital World” [2]. Hartley convincingly argues that in the digital age it is ever more important to have an understanding of what we

One of the key findings was that engineers often do not see the value that MBAs bring to the table, which brought a wry smile to my face.

data preparation and in feature engineering, and it seems naïve to think that my MBAs are truly prepared to tackle them on their own. As I was chewing on this, my friend and colleague J. D. Schramm pointed me to an insightful blog post [1] co-authored by Ali Chaudhry and Rishabh Bhargava. Motivated by their own experiences, the authors first candidly describe the way in which MBAs and engineering students regularly fail to connect with one another, which in turn had led them to conduct focus groups and distribute surveys to try to better understand this phenomenon. From here, they summar ize the results of their analysis and then conclude by making some recommendations about how to improve these cross-campus relationships. One of their key findings was that engineers often do not see the value that MBAs bring to the table. Reading this brought a wry smile to my face. It has been more than 25 years since I finished graduate school in engineering, and more than a dozen years since I started teaching MBAs. Nevertheless, much like the engineering students that Chaudhry and Bhargava had interviewed, I still seem to have a subconscious sense that MBA skills are “easily learnable,” whereas technical knowledge and skills are fundamentally difficult and out of reach for most mere mortals, I realized that it was probably this bias that was at the heart of why I was lamenting the value of the papers that my MBA students had written for my data mining class.

want technology to do for us, and why. “We must embrace technology,” Hartley opines, “yet not forget the liberal arts that give meaning and explain why, not just how, we build.” Describing many of today’s (and tomorrow’s) leaders, Hartley asserts they must “know enough about technology to partner with techies to execute their vision. Their comparative advantage is in their ability to identify problems, and ask questions...” This is exactly what had prompted me to assign my data mining students to write their essays, and more generally why I believe it is valuable to teach analytic thinking and skills to MBAs. While reading Chaudhr y and Bhargava’s blog post and Hartley’s book, I felt myself breathing an involuntary sigh of relief and feeling an unexpected sense of validation, not only for the data mining term paper experiment, but also for my vision for a larger project that I am working on. Long ago, I lamented that, “The nonmathematical factors that help determine success [of analytics projects] are rarely discussed by our journals, our textbooks and our courses. And that is both a shame and a disservice” [3]. A decade later, after being tenured and promoted to professor and rapidly approaching retirement age, I have finally decided to do something about this. In 2017, I began writing a book that focuses on the many human interactions

Like my data mining students writing their term papers, I too am definitely out of my comfort zone. that take place while trying to make mathematical models work in innovative ways. The book is intended for students who are either preparing for or already working on real-world analytics projects as part of their academic training, for instructors who are responsible for supporting them on these projects, and for newly minted analytics professionals looking to gain a broader perspective. My goal is to supplement the readers’ own first-hand experiences by providing my own perspective in my own voice, informed by my own career adventures.

The book’s format is a cross between a textbook and a collection of short stories. Much of the content is presented in the form of essays, with many of them built around stories. This is a deliberate choice. This is partly because of my desire to get these messages to stick (as Rudyard Kipling famously said, “If history were taught in the form of stories, it would never be forgotten.”) and partly because the act of telling these kinds of stories can reveal the key ideas and issues more effectively than bullet points and jargon. As yet, there is no definitive publication date for this book. I managed to get a good start on the project during my sabbatical from teaching last spring, but I still have a lot of work to do. Moreover, while I have a fairly concrete outline, it still feels much of the concrete is still wet,

EDITOR’S NOTE: An earlier version of this article appeared in Analytics magazine.


waiting to be spread and shaped. Like my data mining students writing their term papers, I too am definitely out of my comfort zone. ORMS Vijay Mehrotra ( is a professor in the Department of Business Analytics and Information Systems at the University of San Francisco’s School of Management.

brought to you by

Participate in the 2018 Syngenta Crop Challenge in Analytics brought to you by the Analytics Society of INFORMS and use your analytical and mathematical modeling skills to help solve a real-world challenge: growing more food to meet the needs of a growing population. Enrolled participants will receive a real-world data set and will be charged with developing a unique equation or methodology that addresses a research question relating to crop productivity.





February 2018


ORMS Today

| 17


By Heine Krog Iversen

Hungry for data, thirsty for time Awash in data and opportunities: Why data professionals must move up the corporate food chain. Organizations of all sizes and types are awash in data possibilities, yet most of them cannot capitalize on the potential for a variety of reasons. The good news, however, is that with the right decisions and focus, these possibilities can turn quickly into realized opportunities. Business users in these organizations are hungry for comprehensive, contextual and timely data on which to make decisions. The process of “idea creation, data gathering, decision” is a motivational force for economic development in all organizations. The key to effectively implementing this is the smooth democratization of access to data. Can, in one fell swoop, an organization satisfy its newly enabled users’ data hunger while simultaneously quenching information technology’s (IT) thirst for time? That’s the challenge. As organizations strive for data agility, breaking down internal barriers to data democratization is essential.To stretch the metaphor, dynamic organizations must be able to simultaneously eat and drink. Data democratization and the ensuing benefits are not an episodic or ephemeral phenomenon for the organization; instead, they are fundamental and constant elements of future success. Data infrastructure technology has come of age and is ready to be deployed and exploited in the service of data agility. It is imperative for all organizations to create a plan and chart a clear course of action. It’s time to banish both hunger and thirst. Most C-level executives by now have recognized that becoming a datadriven company is a priority to their 18 | ORMS Today


February 2018

success and that it plays an instrumental role in increasing profit and growth. As technology continues to push the speed of change, we can be assured that this speed of change will continue to increase even faster, and will put even greater pressure on companies to adjust. The amount of IT systems in place is growing, and the average lifecycle of each system is being lowered dramatically. For instance, more and more organizations are trying to keep up by implementing scrum and agile approaches to their implementations. Another way to look at this is that business organizations are putting pressure on IT, and getting more work accomplished in less time demands having data as needed, when needed. Businesses also mandate that new systems are put

into the landscape, forcing IT to know and understand these systems at the speed of light. If IT can’t deliver, companies just bypass IT and turn to cloud offerings. All these new systems and constant change of systems puts pressure on the way organizations build and maintain a governed platform for data analysis.When you add in big data, IoT and all the social data that flows around the business, we can now see that companies are looking for help to provide organization, quality and cost reduction around these systems. The need for speed and the current increasing workload forces data professionals to get out of their comfort zone. Companies that continue to operate as they have over the last 20 years will not survive. Democratizing Data Analysis and Decision-making Big data and social data, along with other data systems and applications, are driving the intelligent enterprise movement. For example,Twitter, Facebook and LinkedIn are now an integrated part of almost every organization’s marketing system. Organizations need to monitor what people are saying about them (the unstructured part), and they need to monitor their followers and how this varies based on what they are doing. It is also important to connect all this data to the

Can an organization satisfy its newly enabled users’ data hunger while simultaneously quenching information technology’s thirst for time? Image © Sean Prior |

Current research also shows that speed in delivering new IT systems or new platforms for data analysis is becoming more important than trying to cut costs.

Apply for 2018

financials to measure the cost per follower or what is called “the structured part.” Data analysis is now becoming a top priority in selecting go-to-market strategies, as well as in the big picture strategy to obtain market leadership. Therefore, companies expected to climb to the top are those that combine internal and external data. Going forward, IT budgets are expected to be taken over by the chief marketing officer or the chief financial officer.This will give IT less impact regarding decisions on what IT will be used, yet IT will have to run and support all these systems. Meanwhile, numerous recent published reports outline

in detail the emergence of data discovery, and why organizations must now consider a governed data discovery hub as their central platform for democratizing data analysis and decisionmaking. Current research also shows that speed in delivering new IT systems or new platforms for data analysis is becoming more important than trying to cut costs. If you believe the proposition that speed is paramount for survival and growth (and I do, by the way), then the only way forward is to empower business users and liberate IT, which can easily be accomplished with data democratization (as long as governance is ensured). Numerous companies are now leveraging data democratization platforms to improve operational performance by saving time, making better decisions and redeploying staff for more strategic, bene-

fit-oriented initiatives. Further, companies are seeing better data-driven decisions by improving the line of business with big data and social data initiatives. Having gover ned and localized data discovery at the fingertips of all business users plays an essential role for organizations trying to lower costs, improve efficiency and enhance data quality. Today, with the deluge of data and the advent of big data and social data, it is imperative to re-engineer the enterprise in this fashion. Companies that do have a much greater chance at gaining a competitive advantage and even survival, while those that don’t risk dying of hunger or thirst. ORMS Heine Krog Iversen (hki@timextender. com) is the CEO of TimeXtender, a software vendor dedicated to democratizing access to corporate data through Discovery Hub, and the largest provider of data warehouse automation software for the Microsoft SQL Server.

Apply to win this prestigious practice prize that rewards professionals who devise innovative analytical methods, utilize those methods in a verifiably successful O.R./analytics project, and describe their work in a clear, well-written paper. Two-page abstract is due by May 1, 2018. This top INFORMS practice prize spans all O.R. and analytics disciplines and application fields. Any work presented in an INFORMS section or society practice-oriented competition is eligible as long as the work did not result in a published paper. The Wagner Prize competition is high-profile, with its own track at the INFORMS Annual Meeting. Presentations are widely distributed via streaming video. Finalist papers are published as a special issue in INFORMS respected practice journal Interfaces. Last year’s competition was held at the INFORMS Annual Meeting, October 22–25, 2017, in Houston, Texas. The first-place prize will be awarded to Lehigh University and Pennsylvania Department of Corrections at the Edelman Gala during the April 2018 Conference on Business Analytics and O.R. in Baltimore, Maryland. Don’t miss your chance to win this illustrious award for 2018.

Daniel H. Wagner February 2018


ORMS Today

| 19


When good A.I. goes bad Validation is the key to avoiding algorithmic disaster. By Joseph Byrum 20 | ORMS Today


February 2018

About as many people say artificial intelligence is going to destroy humanity as say it will solve all of our problems. Perhaps it’s more likely that the outcome will lie in between the two extremes – society as a whole will benefit, but it won’t be perfect. On the optimistic side of the scale, there’s plenty of reason to be upbeat about where A.I. projects are headed.

Face-off: The legal and ethical questions surrounding artificial intelligence. Image © Kheng Ho Toh |

We know from the field of operations research that algorithms can be effective in optimizing just about every critical business process. Adding the power of machine learning could be the key to popularizing O.R. in the industries that do not currently take advantage of what it has to offer, as it takes a sizable investment in time and money to develop customized O.R. tools. Now imagine a software tool that absorbs a company’s operational data and automatically models alternative scenarios to come up with suggestions for better, more efficient processes. The system learns as it goes, as opposed to being set up with pre-written equations. Such an A.I. system would open the potential for faster and more certain results while lowering the barrier to entry for startups and smaller firms that may currently think of O.R. as a luxury they can’t yet afford. Now, give the machine-learning algorithms some autonomy to adjust to conditions in real

time, and you have an A.I. system that can make a realistic contribution to business efficiency. Even in the nonprofit world, they could assist researchers in analyzing massive data sets in scientific fields, combing through the vast emptiness of space to discover new planets, for example [1]. More A.I. Control, More Room for Error But the more control A.I. systems have, the more room there is for error. A.I. systems are essentially software executed on computer hardware, both of which suffer from the inherent limitations of programming. It’s impossible for us to anticipate all the ways in which something might go wrong. The year began with the news that, for over a decade, computer chips from the main suppliers have been vulnerable to highly sophisticated computer code exploits known as Meltdown and Spectre. Nobody noticed the potential for these flaws because the chips have become so complex. February 2018


ORMS Today

| 21

Analytics Ethics

Contrary to the plot of

sci-fi blockbusters, a self-aware,

evil A.I. purposely aiming at pedestrians

is just as unlikely as

one launching nuclear

Intel’s first processor, introduced in 1971, had a manageable number of transistors – 2,300 [2]. Now the company packs 100 million of them in a square millimeter [3], and Intel is well on the way within the next decade of selling a chip with as many transistors as there are neurons in the human brain (100 billion) [4]. Unlike the brain, which evolved over the course of hundreds of thousands of years, those computer chips haven’t been around for very long. If the hardware can go bad, what happens to the decisions of an A.I. that depend on the silicon functioning properly? The legal and ethical questions surrounding A.I. are perhaps most fully explored in the context of self-driving cars, yet even there we seem to lack clear answers. Consider what happens if an autonomous car decides to swerve into a car in an adjacent lane instead of striking a pedestrian crossing the road. Who’s responsible for the crash? The self-driving car’s owner, the pedestrian or the developer of the autonomous technology that made the choice to swerve? Evil A.I. Unlikely Contrary to the plot of so many sci-fi blockbusters, a self-aware, evil A.I. purposely aiming at pedestrians is just as unlikely as one launching nuclear missiles. Far more mundane errors are likely to trouble advanced A.I. programs, because no matter how perfect and benign the underlying algorithms, and how flawless the computing hardware might be, things can still go very wrong. An instructive example can be found in the fate of the Mars Climate Orbiter. NASA launched this probe in 1999, but the orbiter never actually reached the red planet. It didn’t blow up on the launch pad or meet a spectacular demise. Instead, the probe quietly went missing after having traveled more than 400 million miles. It turns out that the engineers at mission control failed to convert readings from pound-seconds into the metric equivalent, newton-


REFERENCES 1. 2. 3. 4. 5. 6. 7. “Trading Program Ran Amok, With No ‘Off’ Switch,” The Wall Street Journal, by Jessica Silver Greenberg, Nathaniel Popper, and Michael J. De La Merced, Aug. 3, 2012. https://dealbook. 8.

22 | ORMS Today


February 2018

seconds [5] before feeding that data into the guidance system. NASA’s internal assessment confirmed that there was nothing wrong with the orbiter itself. Instead, the “verification and validation” process for the project failed to include the software that controllers on the ground were using. The failure to double-check their work cost the agency $125 million [6], not counting the toll of the late-night comedians joking about NASA’s ability to do simple math. Back on earth, an automated trading program at Knight Capital Group “ran amok” [7] in 2012, losing $460 million in what is surely the most expensive cut-and-paste error of all time. Once again, the algorithm itself was fine, but in the process of setting up new code on the company’s servers, a technician accidentally allowed an obsolete bit of code to go live. That code took unwanted market positions worth $6.6 billion. The Securities and Exchange Commission [8] provided insight into what went wrong. Investigators later determined that Knight’s system failed to set trading limits that would have stopped, or at least set off alarms, when the system first began placing massive orders. More importantly, there was no policy in place to double-check the computer code before it went live. A simple test of the system would have revealed the fatal flaw. As happened with the NASA project, validation was the first thing jettisoned in a high-pressure environment.The teams involved likely felt a need to cut corners to get their job done faster. This is, after all, a universal temptation, even in the field of O.R.The brilliance of a well-written algorithm naturally grabs all of the attention, but it’s difficult to get excited about validation, the step that ensures the data are accurate and the results are correct. The lesson as we develop increasingly powerful A.I. solutions is that the A.I. can never substitute for due diligence. The more we turn over decisions to A.I., the more we must ensure that procedures are in place to catch the mistakes made not by the A.I., but the humans who set up and operate the systems. The verification and validation steps must never be skipped, and managers need to make sure that this corner isn’t cut. A.I. systems will still make mistakes, and some of them may even create spectacular headlines. Don’t be surprised when it happens to see a rush to blame A.I. Just know that there’s a good chance that the real cause was a lack of validation. ORMS Joseph Byrum is the chief data scientist at Principal Financial Group. He is a former research and development executive at Syngenta. Connect with him on Twitter@ ByrumJoseph.

What’s Your StORy? John Milne Associate Professor, Clarkson University INFORMS member since 1985 What advice would you give to your younger self? Volunteer for activities that are important and interesting—even though they’re not your job. You’ll learn more, enjoy it, make contributions, and you’ll often be recognized for it. Also, working half-days on Saturdays will boost your productivity 20-30% (due to uninterrupted time) and you’ll still have plenty of weekend time for other matters (including relaxing/exercise). Tell us about your experiences at IBM. In 1984, I joined IBM’s semiconductor manufacturing business in East Fishkill, NY, to help design and develop a new software system for production control. Mostly, it was trying to understand what the users were doing manually and then designing software to automate it or—when their method was ill-defined—developing a method that seemed logical. It wasn’t until a decade later that I first used linear programming for planning IBM’s semiconductor division supply chain. By that time, the problems had grown more complex, and key members of the user community were comfortable using computers to make calculations that they could not perform personally on their hand-held calculators. In the years that followed, computing power and user comfort with it continued to increase. We saved IBM hundreds of millions of dollars by re-engineering the semiconductor supply chain planning and decision-making process. You have more than 40 U.S. patents—tell us your favorite! Material requirements planning (MRP) systems create plans that can be infeasible indicating actions that should have been done in the past. Though infeasibilities are undesirable in most contexts, in the context of a site MRP run, the site’s delivery commitments have already been made so the infeasibilities indicate where expediting is required. On the other hand, for a new IBM technology (with double-speed sorted product), linear programming (LP) was more appropriate for wisely allocating assets. I spent weeks trying to figure out how to make MRP smarter or to get an LP to indicate the expediting needs. Neither approach worked. Pressure for me to deliver was growing along with my frustrations. Finally, it occurred to me how to blend both LP and MRP technologies. This led to my first and favorite patent (U.S. Patent 5,943,484).

More questions for John? Ask him in the Open Forum on INFORMS Connect!

Future directions for broader impacts at the NSF

By Sheldon H. Jacobson, Jerome F. Hajjar, Andrew Johnson, Erick Moreno-Centeno and Siqian Shen

24 | ORMS Today


February 2018

On May 26-27, 2016, a group of 30 researchers and administrators in the areas of mechanical, industrial and civil engineering met to discuss a game plan to uncover how investments made by the National Science Foundation (NSF) in these areas can be enhanced via “broader impacts” contributions. Through a series of presentations on broader impacts contributions by academic thought leaders in these engineering research communities, plus a vibrant discussion on the issues related to broader impacts criterion, several points came across that would add value to NSF’s basic research investments. “Intellectual merit” and “broader impacts” criterion both require significant attention if NSF is to continue to contribute to the economic and social well-being of the United States, and advance the nation’s worldwide position of preeminence and leadership in science, technology, engineering and mathematics. Intellectual merit and broader impacts both have a significant impact on the development of human capital. In a society that demands a growing number of highly trained, technology literate and skilled people, the need to grow and nurture such talent domestically, across all facets of society, is critical to our nation’s economic wealth and societal stability.

Grass-Roots Effort A grass-roots effort must be used to organically grow the emphasis on and enhance the value provided by broader impacts contributions within the NSF target community of current and future principal investigators (PIs). Dictating such an emphasis from the top at NSF will only have limited value and a short life span. At the same time, the implementation of evaluating broader impacts must be done at the review panel level, through communication by program directors to their panelists, so that proposal evaluations appropriately reflect innovations captured in both intellectual merit and broader impacts, with funding decisions made accordingly. Indeed, researchers must be incentivized to embrace broader impacts as a critical element of a proposal, and reviewers must be educated to embrace broader impacts as a substantive component in the evaluation process, not just an add-on to the intellectual merit contributions. Intellectual merit and broader impacts cannot be viewed as either/or entities, but rather, complementary and synergistic contributions. The objective of the workshop was to outline a roadmap for enhancing broad impacts activities in NSF grants within the Division of Civil, Mechanical and Manufacturing Innovation (CMMI). In order to elicit ideas from the workshop participants, several questions were posed. The questions served to initiate discussion on how innovative ideas can be facilitated from the research community and how institutions can work with PIs to realize more significant broader impacts by leveraging institutional resources. Key Points from the Workshop The key points that came across during the workshop are as follows: 1. Encourage NSF to provide a framework for thinking about broader impacts that avoids a list of prescriptive activities. Such a framework will enable PIs to define more clearly their project’s broader impacts and will make reviewing and evaluating broader impacts more consistent. We propose that broader impacts be classified along three dimensions: (i) immediacy to the intellectual merit impact, (ii) societal benefit, and (iii) outcome time horizon. Immediacy to the intellectual merit impact can be intrinsic (high immediacy), direct (moderate immediacy) and extr insic/far-reaching (low immediacy). Societal benefit can be in the domain of information/communication, global leadership, people, economic, health and national interests. The

Can investments made by the National Science Foundation be enhanced via “broader impacts� contributions?

The objective

outcome time horizon can be classified as short term (concurrent to grant period), medium term and long term.The classes on all three categories are neither exhaustive nor rigid; broader impacts may fall outside this structure or may fit in several.

of the


2. Encourage NSF to clarify and emphasize to all of its stakeholders that broader impacts can be the impact of the intellectual merits of the proposal and broader impacts are not necessarily achieved through extra activities beyond the intellectual merits activities. NSF does not require proposals to have broader impacts activities beyond the broader impacts intrinsic to the intellectual merits.

was to

outline a roadmap for

enhancing broad impacts

3. Improve communication of broader impacts between NSF and its stakeholders (program managers, reviewers/panelists, principal investigators, Congress and the general public). Specifically, both the evaluation of broader impacts in proposals and the reporting of broader impacts to the general public should be emphasized. 4. According to the type of societal benefit they bring, broader impacts can be classified as: information/communication, global leadership, people, economic, health and national interests. a. Information/communication: Any activity that brings new insights onto existing and February 2018

activities in

NSF grants within


ORMS Today

| 25

Broad Impacts at NSF

“Intellectual merit” and

“broader impacts” criterion both

require significant attention


if NSF is

continue to contribute to

to the




social well-being of the

United States.




26 | ORMS Today


anticipated societal issues or promotes and protects professional standards. Such information may lead to better informed societal decisions, which may in turn enhance economic and societal well-being. Also, any activities that promulgate societal benefits and/or inform population sectors that improve quality of life and enhance public literacy. This may include communication and outreach via social media, websites, blogging, videos and traditional media (radio, newspaper, television). Global leadership: Any activity that strengthens the United States’ world leadership (e.g., economic, technological, productivity, sustainability). The broader impacts in this category generally, but not necessarily, will be a direct or indirect result of broader impacts achieved in other broader impacts categories. An example of broader impacts in this category is that of maintaining and growing leadership in strategic technologies such as information systems. People: Any activity that educates people or increases diversity in the human pool of skills resulting from the research. For example, this can be bringing STEM activities in K-12 environments, enhanced opportunities for underrepresented populations or training a more capable workforce. Any activities that enhance and enrich human capital fall within this classification, including the impact of policies on the lives of U.S. citizens. Economic: Any activity that directly or indirectly leads to economic value being created or facilitating the creation of economic value. For example, this can be enhanced productivity, job creations or cost reductions. Health: Any activity that enables the development of new health devices and/ or medicines. This includes any activity that directly or indirectly leads to a better quality of life or societal well-being as well as activities that bring new insights onto existing or anticipated health issues. Such contributions are not of a clinical nature, but rather, include societal impacts related to health. National interests: Any activity that brings advancements aligned with national interests, such as national security (e.g., reduces or eliminates the threat of attacks via nuclear, biological and chemical weapons) or national stability (e.g., enhances the viability and

February 2018

stability of national systems including energy supplies, transportation systems, financial markets and/or the environment). 5. Emphasize that the proposed classification framework is neither prescriptive nor exhaustive. Specifically, the aforementioned classes are neither exhaustive nor rigid; broader impacts may fit in several classes or in none. As such, this classification framework is only meant to: (i) facilitate researchers to frame the broader impacts of their project and to think of additional broader impacts that they may not realize their project encompasses; (ii) make explicit the diverse and vast broader impacts attained by the activities funded by NSF. 6. Invigorate efforts to track, document and disseminate the medium- and long-term broader impacts. A first step is to allow PIs to append/update the final project outcomes/abstracts with broader impacts. A second step entails requiring PIs to report the broader impacts achieved five or 10 years after the project end date. Moreover, NSF should implement a highly prestigious and highly broadcasted/disseminated broader impacts award.There can be one such yearly award for each type of societal benefit broader impacts category, one for each immediacy to intellectual merit broader impacts category, and one for each outcome time horizon category, as well as career-long awards (with visibility similar to the PECASE awards) targeting researchers with lifelong achievements of broader impacts.These awards will not only incentivize PIs to report their broader impacts, but also help to disseminate them. 7. Intrinsic and direct (typically short-term) broader impacts are inextricably linked to intellectual merit; in contrast, extrinsic/far-reaching broader impacts are not. The workshop participants believe that a possible approach to incentivize extrinsic/far-reaching broader impacts activities would be via a two-phase procedure. This procedure would work as follows: First, evaluate and select proposals based on strong intellectual merit, as well as intrinsic and direct broader impacts. Subsequently, PIs of funded proposals can propose additional extrinsic/far-reaching broader impacts activities and request an additional funding supplement. The size and review procedures for such broader impacts supplements would be analogous to those of the REU supplement.This two-phase approach will allow PIs to have more information, time and resources to pursue more difficult and substantial tasks in terms of innovating educational forms and reaching out to general public. Moreover, since these activities are funded,

it will be easier to track their results and also to hold PIs accountable. The workshop provided an environment for spirited discussion and sharing of ideas to promulgate the value of broader impacts activities. This resulted in the aforementioned ideas and guidelines to facilitate quantum leaps forward to enhance NSF’s investments in basic research. The outputs from this workshop serve as recommendations. NSF will ultimately decide which to adopt and implement. With these recommendations in hand, NSF is now positioned to take bold steps to elevate the expectations for broader impacts activities in the proposal that they support. The academic research community can respond by using their creativity and talents to enhance their efforts on broader impacts, and demonstrate that intellectual merit and broader impacts can both be transformative, substantive and impactful. ORMS



18 INFORM 20 S

Sheldon H. Jacobson is a founder professor in Computer Science at the University of Illinois. He has a longstanding commitment to broader impacts, serving as the director for operations research at the NSF from 2012-2014.

J u ne 1 7 – 2 0

Jerome F. Hajjar is the CDM Smith Professor and department chair in the Department of Civil and Environmental Engineering at Northeastern University. He is committed to fostering a broad understanding nationally of the seminal impact for society of NSF-sponsored research and education. Andrew Johnson is an associate professor in the Department of Industrial and Systems Engineering at Texas A&M University and a visiting associate professor at Osaka University. His international collaborations have helped him understand the value of broader impacts not only locally, but also globally. Erick Moreno-Centeno is an associate professor in the Department of Industrial and Systems Engineering at Texas A&M University. Siqian Shen is an associate professor in the Department of Industrial and Operations Engineering at the University of Michigan. She is currently an associate director for the Michigan Institute for Computational Discovery and Engineering (MICDE).

ACKNOWLEDGEMENTS The authors thank Dawn Tilbury for her contributions to this effort. This article was written with support from the National Science Foundation (CMMI-1629955). Any opinions, findings, conclusions or recommendations expressed in this material are those of the authors and do not reflect the views of the U.S. government or the National Science Foundation.


A Better World through O.R., Analytics, and AI At the 2018 INFORMS International Conference you will discover advances in deep learning and smart artificial intelligence from the world’s leading innovators, learn the latest technological advancements and industry trends from a global line-up of experts, and interact with business leaders, influential technologists, & data scientists leading the deep learning revolution.

Why You Should Attend: • Expand Your Professional Network • Explore Emerging Trends in AI • Share Latest Research in Analytics, O.R., and AI

Abstract Submission Deadline: March 15, 2018

Register Today & Save!

February 2018


ORMS Today

| 27

The digital marketplace is rapidly changing the global music industry. Image Š Pavel Ignatov |

O.R. and music generation The role of operations research in state-of-theart automatic composition systems: challenges and opportunities.

By Dorien Herremans and Elaine Chew

28 | ORMS Today


February 2018

Operations research (O.R.) is a diverse field centered on the modeling and solving of complex problems. The robustness of O.R. techniques is evidenced in their successful applications to problems as diverse as finding optimal phone keyboard layouts [26] and modified driven equilibrium fourier transform sequence optimization to improve 3-D brain imaging scans [7]. Apart from addressing problems of industrial utility and health, operations research has also been applied to problems farther afield in art and design such as architectural layouts [8] and the finding of musical solutions that satisfy compositional constraints [28, 10]. It is the last category of problems that forms the focus of this article – the role of operations research in state-of-the-art automatic composition systems, the challenges and opportunities. In particular, we highlight the contributions of O.R. modeling in the context of the MorpheuS project [30, 15]. For a more complete overview of music generation systems, the reader is referred to Herremans et al. (2018) [17].

The Digital Music Economy The digital marketplace is rapidly changing the global music industry. In 2014, the industry’s digital revenue increased by 6.9 percent, reaching a total of $6.85 billion [9].Vast amounts of music-related apps have surfaced for smartphones. These easy-to-use and seemingly simple apps are powered by clever algorithms, often inspired by techniques common to O.R.; cases in point include Shazam and Spotify. In this article, we aim to highlight an up-and-coming area that has only recently gained popularity: automatic music generation. Automatic music generation serves many purposes. The objective is not to replace composers – the algorithms still rely heavily on human-conceived designs and are nowhere near producing high art music – but serve the purpose of creating changeable music in scalable ways for more mundane recurring scenarios. Imagine a world where small independent video makers can access large amounts of copyright-free and bespoke music for use as background music to their videos, or where a user can generate a novel soundtrack to accompany their vacation photo slideshow for online sharing. What if a computer game could have different and new music that follows the suspense each time the player enters the game? It is scenarios like these that have provided researchers with the motivation for developing methods for automatic music generation. The idea of automatically composing music existed long before the first computers were built. In the words of Ada Lovelace, widely regarded to be the world’s first programmer, as she wrote about the general-purpose computing machine: [The Engine’s] operating mechanism might act upon other things besides numbers [...] Supposing, for instance, that the fundamental relations of pitched sounds in the signs of harmony and of musical composition were susceptible of such expressions and adaptations, the engine might compose elaborate and scientific pieces of music of any degree of complexity or extent.” ­ Lovelace, 1843 [18]

Today, several techniques exist for generating music that sounds reasonable at the local scale. The big challenge that the field faces is that of creating music that possesses long-term structure and large-scale coherence. We believe that looking at music generation as an optimization problem offers a viable framework for solving this problem.

Optimization-Based Approaches to Music Generation Composing music can be construed as a constraint satisfaction or combinatorial optimization problem. Composers such as Stravinsky have likened the composition process to one of following self-imposed constraints: “My freedom consists in my moving about within the narrow frame that I have assigned myself for each one of my undertakings. ... The more constraints one imposes, the more one frees one’s self of the chains that shackle the spirit.”

Composing music can be construed as

a constraint satisfaction

– Stravinsky (as quoted in Strauss [27])

The task is then to find the right combination of notes that best satisfies these constraints. But how does one judge if a piece of music is good, or if one solution is better than another? The late music theorist Edward Cone has written that “a good composition manifests its structural principles” and that most music criticism “sets about discovering the standards implied by a given work and testing how well it lives up to them” [6].This theory thus offers a way to assess the quality of a composition, but evaluation – how we construct an objective function that effectively measures the quality of music –­ remains a central challenge in music generation.


combinatorial optimization problem.

Quantifying Music Quality There are three main ways to evaluate a musical piece. The first is via human evaluation. Arguably the best way to find out if a piece sounds good is to test it in front of a live audience. Unfortunately, this technique does not scale well and would be impossible to embed in a quantitative algorithm as it would cause huge delay. This problem is often referred to as the human fitness bottleneck [2]. Next, we could turn to music theory to obtain a set of rules that defines what is permissible in a given style. Music generation algorithms that rely on rules focus primarily on counterpoint, a style of music popular in the Renaissance period that has multiple harmonically interdependent voices [10, 23]. Counterpoint is one of the most formally defined musical styles; for most other genres of music, no such well-behaved set of rules exist. A third, more robust technique is to use machine learning to construct the objective function. Since the 1950s, Markov models have been used to capture statistical properties of musical pieces and genres [24, 3]. While these models can capture many aspects and features of music, their traditional sampling methods (e.g., random walk, Gibbs sampling) do not provide a method for generating structure and thus the resulting pieces often lack February 2018


ORMS Today

| 29

O.R & Music Generation

Figure 1: Multiple embedded helices in the spiral array representing pitch classes, major and minor chords (triads), and major and minor keys (Figure reproduced from Chew 2014, p.59 [4]).

long-term coherence. More recently there has been interest in using deep neural networks for composition [16,17], yet the problem of constraining long-term structure remains. Herremans et al. [12] developed a method to integrate Markov models in the objective function of a metaheuristic.The approach allows for the resulting music to be evaluated based on a machine-learned model, yet supports the use of hard constraints to fix a larger structure. In our MorpheuS system [13, 14], this approach is further expanded to tackle complex polyphonic music and automatically detected recurring patterns.We also developed a way to combine machine learning and music theory to construct a new type of objective function based on musical tension.

Representing Music In order to work with music in a quantitative way, we need a mathematical representation. Typically, music consists of a number of notes, each with properties such as pitch (often expressed as a MIDI pitch), duration, start time and beat number. In the MorpheuS

project, the music generation problem is re-cast as one of finding new pitches that satisfy certain structural constraints given a rhythm template extracted from an existing piece. Our goal is thus to morph a pre-existing piece into a new composition. Almost all the music that we hear is tonal, i.e., adhering to a system of hierarchical pitch relations. The tonal properties of a piece of music and how they evolve over time shape the listener’s perception of musical stability and tension.The tension profile of a piece is one of the main defining characteristics that imbues upon the piece long-term structural, which we use as a structural constraint in the MorpheuS project. This tension profile is calculated based on the spiral array model for tonality [4]. The spiral array consists of multiple embedded helices (see Figure 1), representing tonal entities at three hierarchical levels: pitch classes, chords and keys. The pitch class helix is effectively the line of fifths (most simply explained as pitches having a simple 2:3 frequency ratio) wrapped around a cylinder so that pitch classes align vertically every four quarter turns. Chords and keys are each generated as convex combinations of their defining elements as shown in Figure 2. A real-time implementation of the spiral array and its tonal analysis algorithms has been used in live performances [5] to visualize tonal relationships in a music piece (see Figure 3). Tension Ribbons Leonard Meyer [22] ascribes the emotional content of music to the composer’s choreography of

Figure 2: Spiral array pitch class, chord, and key representations: chords are weighted combination of their component pitches, and keys of their defining tonic (rooted on the first scale degree), subdominant (rooted on the fourth scale degree), and dominant (rooted on the fifth scale degree). (Figures reproduced from Chew 2014, p.54-55 [4]).

30 | ORMS Today


February 2018

Figure 4: Joplin’s German Sixth chord mapped to the spiral array; the chord in the key of Eb major consists of the notes Cb, Eb, F# and A. Figure 3. Elaine Chew performing with the MuSA_RT app [31] at the New Resonances Concert of the 2012 International Computer Music Modeling and Retrieval Conference.

expectation. Expectation suspended elicits tension, and expectation fulfilled leads to resolution of that tension. The MorpheuS system aims to recreate the palpable tension that music generates; to this end, we have developed a model that captures tonal tension based on the spiral array [13]. The spiral array tension model first segments a piece into equal length subdivisions and maps the notes to clouds of points in the array. Three aspects of tonal tension are captured using these clouds: the cloud diameter measures the dispersion (an indicator of dissonance) of note clusters in tonal space; the cloud momentum measures the movement of pitch sets (change in tonal context) in the spiral array; and, tensile strain measures the distance between the local and global tonal context. Each attr ibute can be visualized as tension ribbons on the score (see Figure 5). Figure 4, shows a well-known tense chord in the spiral array: the German Sixth. It is noticeably spread out in the tonal space. The German Sixth chord occurs on the last beat of the second bar of Scott Joplin’s “Binks’ Waltz” (Figure 5). When this chord occurs, there is a noticeable increase in the tension ribbons representing cloud diameter


and cloud momentum in the spiral array model. The tensile strain starts to increase slightly earlier, adding to the buildup of tonal suspense leading up to the German Sixth chord.


unfamiliar music

Patterns in Music Repetition forms another important aspect of music structure. Recurrent themes and motives are reinforced with each repetition and stay in the listener’s mind through the listening experience. A study by Margulis [19] revealed that listeners rated unfamiliar music that had repetitions as more enjoyable, interesting and artistic than the original non-repeated version. With repetition, patterns begin to form; Plato alludes to the importance of patterns in music: “I would teach children music, physics and philosophy; but most importantly music, for the patterns in music and all the arts are the keys to learning”

that had

repetitions as

more enjoyable, interesting and


– Plato as cited by Plato & Bloom [25].

Thus, in MorpheuS, we aim to generate music with patterns having the same length and that recur at the same places as the original template piece. In order to generate music with recurring patterns, we need to first find them. Compression algorithms such as SIATECCompress [20] and COSIATEC [21] can be used to find maximal translatable patterns, i.e., the largest subset of consecutive February 2018


ORMS Today

| 31

O.R & Music Generation by Herremans [11] for generating counterpoint music. This original VNS has been extended – with a move that changes multiple notes in a time slice – in order to handle more complex polyphonic music. (a) Cloud diameter

(b) Cloud momentum

(c) Tensile strain Figure 5: Tension ribbons of the German Sixth chord in Scott Joplin’s Binks’ Waltz (1905).

notes that recur either verbatim or transposed by a constant interval. Using the elements discussed above, we can define music generation as an optimization problem. MorpheuS takes as an input a template piano piece. From the exemplar piece, the rhythmic structure is extracted and retained as a template.The algorithm aims to find new pitches for each note in the original piece, and thus morph the piece into a new one. Recurring patterns in the template piece are detected using COSIATEC. These patterns form hard constraints during the optimization so as to enforce perceptible musical motifs and to ensure long-term coherence. The objective of the algorithm is to generate music with a specific tension profile; the user can choose to have the algorithm match the tension profile of the template piece, or follow a user-given tension profile. To solve the multi-objective tension profile matching problem, we have implemented an efficient variable neighborhood search (VNS) metaheuristic based on the algorithm developed REFERENCES & NOTES The complete list of references and notes cited in this article are available online at:

32 | ORMS Today


February 2018

Conclusions Systems like MorpheuS can now compose music with limited global structure, which may well be a starting point for scalable systems for automatic music generation. The output is promising, but still falls short of human creations. Take, for example, the issue of playability. The MopheuS-Haydn piece contains many awkward leaps and hand crossings; although these are not impossible, they do not often appear in human compositions. In the literature, apart from some algorithms for rudimentary fingering [1, 29], quantifying playability (which is far more than fingering) remains an important challenge. Secondly, little is known about how the interactions of local and global patterns generate coherence. Could there be a way to generate viable patterns without relying on a template? Third, the shaping of music is closely tied to body movement, and feelings such as stress and lilt are part and parcel of musical communication.There are currently no known methods for, say, generating a lilting feeling or incorporating a danceability factor, which is not automatic even when the time signature is three-quarters (as in a waltz). Finally, aspects such as having the time to breathe and to respire are essential components of human and thus musical communication. How does a composer modulate, beyond tonal tension, other factors for moments of release? As we consider and tackle these challenges, we find ourselves faced with the deeper question of what it means to be human. ORMS Dorien Herremans is an assistant professor at Singapore University of Technology and Design and has a joint-appointment at the Institute of High Performance Computing, A*STAR. Before that she was a fellow at the Centre for Digital Music at Queen Mary University of London. She received her Ph.D. in applied economics on the topic of “Computer Generation and Classification of Music through Operations Research Methods.” She graduated as a commercial engineer in management information systems at the University of Antwerp in 2005. Elaine Chew is professor of Digital Media at Queen Mary University of London’s School of Electronic Engineering and Computer Science, where she is affiliated with the Centre for Digital Music and serves as its director of Music Initiatives and co-lead of the Cognition, Creativity and Expression research theme. Her scientific research centers on the design of mathematical and computational tools to model, analyze, visualize and manipulate music structures.


EXHIBIT & SPONSOR AT THE 2018 ANALYTICS CONFERENCE IN BALTIMORE, MD Join analytics practitioners and experts from a variety of industries and sectors. For additional details, contact Olivia Schmitz at or visit: Baltimore, Maryland


April 15-17, 2018

Insurers are increasingly calling for screening, and where appropriate, treatment of mental health issues by primary care clinicians. Image © lightwise |

Measuring medical policy changes with O.R. An application to the diagnosis and treatment of depression

By Sumana Reddy and Harrison Schramm

34 | ORMS Today


February 2018

How can advanced statistical methods be made accessible and gain insights for small- to mid-sized medical practices? For the co-authors, the answer began with a series of conversations about how questions arising from current medical practice could be addressed using analytics.The hope is that readers will see the ensuing project and examples presented here as motivation to consider using data analytics – and the professionals who perform analysis – in their own practices to gain insights improving both management and patient care.The co-authors also hope to demonstrate that this type of analysis is not just a “major medical center” or “large practice” activity, but also is within reach for smaller practices. Methodology The Acacia Family Medical Group (AFMG) data used in this analysis consists of two tables containing appointment data and diagnosis codes. The appointment data consists of approximately 165,000 rows to include “noshows” and cancellations. The diagnosis data consists of dates, ID and diagnosis codes. This data spans the time period 2012-2017; the full, clinically relevant patient history is not available.

Confidentiality of patients is preserved by using a randomly generated identifier for each patient, such as “ZZZZZ0002D” (see Table 1). Confidentiality of doctors is maintained by randomly recoding their name to a letter, i.e., Dr. “A.” The data analysis was performed in the statistical programming language R, with heavy use of the packages that make up the socalled “tidyverse” [1]. Depression The use case of this study focuses around depression, as denoted by the following two International Classification of Diseases (ICD-9) codes [2]: • 296.2: major depressive disorder, single episode • 311: depressive disorder not elsewhere classified Our effort focuses on ICD-9 codes; to maintain consistent methodology across the timespan of data, we do not consider current (ICD-10) codes.The switchover between systems occurred in October 2015. One of the objectives of this work was to consider the shift in diagnosis and coding of depression over time. Legislation, particularly the 2008 Mental Health Parity Act [3] and the 2016 21st Century Cures Act [4], allowed for increased access to mental health benefits. Due to the shortage of mental health professionals, insurers are increasingly calling for screening, and where appropriate, treatment of mental health issues by primary care clinicians.We expected to see a gradual change in prevalence of diagnoses, as clinicians adapted from an environment pre-2008 in which a diagnosis of depression was not reimbursable, to one in which recognition and treatment is encouraged. Acacia Family Medical Group AFMG is a medical practice serving the full spectrum of clients in California’s Salinas Valley. It has offices in the cities of Salinas and Prunedale. Salinas and the surrounding catchment area are rural and semiurban, and patients are largely in city government and services such as teaching, as well as agri-business employees from executives to farm laborers. It is a younger than average population when compared against the region, with a high proportion of working-class people.The practice generally employs four physicians and a physician assistant at any given time; more doctors are present in the data set due to individuals joining and leaving the practice. It has participated in all available quality metrics and programs through Medicare and private insurers. Results As a first step, we considered how many patients in AFMG have been diagnosed with depression, by year (see Figure 1). This chart shows an overall increase in the treatment of depression across a stable








Salmonella Gastroenteritis





Gastroenteritis due to food poisoning





Gastroenteritis due to food poisoning





Clostridium Difficile Gastroenteritis





C. Difficile Colitis





C. Difficile Colitis



Table 1. Sample data from analysis. This raw data was grouped by ID to determine a “pattern of life” for patients with specific conditions.

patient population, with particular growth in the 296.2 code. A deeper analysis considers the first time that a patient is diagnosed with depression, the average time between visits (for those who had more than one visit) and the total number of visits (see Figure 2).These are broken out for readability between the 296.2 and 311 ICD diagnoses.

Figure 1: Appointments for depression by year and type.

Figure 2: Patient history by date of first diagnosis. The dates on the X axis are the date first diagnosed, the Y axis represents the average time (in day) between visits, and the size of the point represents the total number of visits. February 2018


ORMS Today

| 35

Healthcare may have broad applicability in extracting other insights from data.

Figure 3: Co-morbidity with depression diagnoses, grouped by epoch.

Figure 4: Initial diagnoses of depression by doctor (doctor names obfuscated).

This approach is useful in several contexts because it allows practitioners to create a “pattern of life” for patients. Once a “pattern of life” is determined, individual cases may be cross-compared, and outliers – doctors or patients that are “off trend” – may be identified. This approach ACKNOWLEDGEMENTS The authors thank Daniel Von Forell of SVMH for his assistance obtaining raw data, as well as the CANA Foundation for its support of this effort.

REFERENCES 1. Wickham, H., 2014, “Advanced R,” CRC Press, Boca Raton, Fla. 2. Centers for Disease Control and Prevention, 2017, International Classification of Diseases – 10th Revision – Clinical Modification; 3. Centers for Medicare and Medicaid Services (CMS), 2008, “The Mental Health Parity and Addiction Equity Act (MHPAEA); 4. Food and Drug Administration, 2016, “21st Century Cures Act”;

36 | ORMS Today


February 2018

Co-morbidity at First Diagnosis We can also consider the difference in co-morbidity, specifically what other conditions are frequently diagnosed when a person is first diagnosed with depression, and how does this change in time? Our hypothesis, based on our practice, is that patients rarely make appointments for visits expecting a diagnosis of depression. It is our further hypothesis that the character of first appointments changed around 2014, because it took some time for the change in policy to “trickle down” to individual practitioners (see Figure 3). Finally, we can consider which doctors are performing initial diagnoses for depression (see Figure 4). A chart like this needs to be read with care; one conclusion is that the doctors with the highest number of diagnoses of depression are more apt to make this diagnosis. A competing hypothesis is that in a family practice, individual doctors may tend to treat groups of patients that are at higher risk for depression. Evaluating these hypotheses is beyond the scope of data and effort for this study. Conclusion and Future Work This article is a demonstration of the use of alternate statistical tools on a data set from a family practice to gain insights into that practice. Our sense is that we have barely scratched the surface as to the insights to be gleaned. Specifically, this or similar analysis could be performed on any ICD code or collection thereof. Our major takeaway for readers is that this sort of analysis is within the technical reach of many practices with a modest amount of external help from professionals in statistics and/or data science. With a small adjustment, these approaches and data sets like this could be applied to finding insights about prescriptions and the “pattern of life” for patients across a variety of questions. Specifically, they could also be used to detect emerging trends in patient needs, as well as internal mechanisms for practices to perform their own quality audits. ORMS Sumana Reddy, M.D., is a family practice specialist in Salinas, Calif. A fellow of the American Academy of Family Physicians (AAFP), Dr. Reddy is board certified in Family Medicine (U.C. San Francisco Medical School) and is affiliated with the Acacia Family Medical Group. Harrison Schramm (, CAP, PStat, is a principal operations research analyst at CANA Advisors. He earned a master’s degree in operations research at the U.S. Naval Postgraduate School.

Call for Nominations Editor-in-Chief, Decision Analysis

Deadline for Nominations: March 31, 2018 The final term of Professor Rakesh Sarin as editor-in-chief of the journal Decision Analysis will end on December 31, 2018. All members of INFORMS are invited to participate in this process to propose a new candidate. Qualifications for the editor-in-chief of Decision Analysis and further information about the journal is available on the website: Please submit comments, recommendations, and nominations (including self-nominations) to the committee by March 31, 2018.


Data Governance

The rise of

self-service analytics As SSA gains momentum, the need for data governance increases in order to drive true business value.

By Paul Brunet

38 | ORMS Today


February 2018

With the rise of big data – and the processes and tools related to utilizing and managing large data sets – organizations are recognizing the value of data as a critical business asset to identify trends, patterns and preferences to drive improved customer experiences and competitive advantage. The problem is, users too often can’t find the data they need to perform desired analytics. Data tends to be buried in different systems or siloed in departments across the organization. This data chaos and uncertainty is costing businesses big money – as much as $3.1 trillion, according to a recent Harvard Business Review study. In addition to the time wasted on searching for the data, individual interpretation of the data through a subjective lens can result in inconsistencies that adversely affect a company’s business. Making Trust a Priority for Reliable Analytics The industry has seen a surge of self-service analytics (SSA) tools such as Tableau and Qlik that enable analysts and non-technical business users to gain insights and drive data-focused initiatives. SSA and business intelligence (BI) empowers knowledge workers and business users to gather desired insights without reliance on IT to run reports. However, investing in analytics tools alone can’t deliver business value. For a SSA tool to do its job, companies need to ensure that the people using the tool can easily access the data they need across the organization – including siloed data living in various systems – and have full confidence in this data to apply it for greater business insight and results. Effectively integrating disparate data from different systems and devices requires a complete understanding of the organization’s “data map” and the data’s journey and relationship to other similar – or sometimes contradictory – data throughout the organization.This is best achieved through data governance. Data governance offers a collaborative framework for managing and defining enterprise-wide policies, business rules and assets to provide all business users with high-quality data that is easily accessible in a consistent manner. By adopting an overall policy through governance, users can determine data inventory, data ownership, critical data elements (CDE), data quality, information security, data lineage and data retention so they have a good understanding of the data across the organization and its meaning.True data governance breaks down data silos so users can find the trusted data they need, collaborate on it, and easily understand it so it’s consumable to drive competitive advantage.This is the new order of data governance today.

Data governance offers a collaborative framework for managing and defining enterprise-wide policies, business rules and assets. Image © marigranula |

Ensuring data trust has become one of the most critical factors to driving successful BI initiatives.When users know they can trust the data, they are more likely to use it for business insight. And this element of trust becomes even more critical when we look at automated analysis through machine learning, a growing trend that offers great business value.The ability to sift through large volumes of data and draw conclusions can move a business forward. But in a business world where the volume of data has become increasingly massive, it’s impossible to manage this without automation. Empowering Data Discovery for Greater Insight Data governance is critical to the success of self-service BI models by providing consistent and reliable data across the organization in a unified form that algorithms can understand.Additionally, business users need to know how to best explore this data without relying on IT’s hand-holding for optimal discovery and insight. Initial user training of analytics tools is essential before getting started on any project. But even for those who possess a good understanding of how to use these tools, agreeing upfront on definitions and KPIs of BI models is essential, as is knowing whether certain reports already exist before duplicating efforts. This is another common bottleneck to leveraging SSA

tools’ full potential – research shows 70 percent of a data analyst’s time is spent on preparing and analyzing data on questions already answered.The more visibility and understanding users have of the data, the more informed decisions they can make regarding which models to explore.

Ensuring data trust

Consumerizing Data Discovery and Analytics via a Data Catalog Deploying a data catalog as part of the data governance solution provides business users with a more strategic and simplified data overview. A data catalog organizes useful collections of data across existing boundaries. Whether those boundaries are systems, organizations or geographies, this cross-boundary visibility drives many of data’s more significant insights.This broader understanding of data through governance and a catalog empowers the user with a consumable data experience for capitalizing on analytics. Users can harness the expertise of data citizens around the organization, too, and use the catalog infrastructure to enable them to share their work.This gives the user a clear idea of which reports already exist and results in more effective BI analytics and reporting. Through a catalog, business users can easily find (or shop) for the trusted data they need from one central location, just as they do on consumer sites such as

one of the

February 2018


become most critical factors to driving

successful BI initiatives.


ORMS Today

| 39

Self-Service Analytics


Amazon. A catalog automatically links business terms from the business glossary to registered tables and columns, and leverages the organization’s agreed-upon vocabulary, providing users with a better understanding of the data’s context. This helps users determine whether that data is a good fit for the analysis in question. And different views of the data provide different aspects, which feeds into the analysis. If the data for which users are searching is not properly cataloged, the self-service tools will not yield valuable results. A data catalog incorporated with machine learning delivers even greater insight and makes recommendations based on past user behaviors and “data purchases,” much like Amazon does for frequent shoppers.A catalog makes it easier and quicker for users to find the data for decision-making, but also enables users to define models earlier in the process.This is particularly helpful for making changes on the fly, as in the case with last-minute requests in definitions or KPIs. To gain full value of SSA tools, the data catalog should support five capabilities: 1. Consumability. Choosing a data catalog and self-service analytics tools with a user-friendly interface is important to business users who may not be tech savvy. Simple drag-and-drop functionality, intuitive mapping and navigation, and easy-to-read help sections are imperative. The data catalog should go beyond structural and usage metadata and provide an easy “data shopping” experience with the complete meaning, lineage and relationships of consistent and trusted governed data for the business user to capitalize on analytics for business advantage. 2. Business modeling. A catalog with out-of-thebox operating models complements self-service analytics tools by providing a flexible structure for consumable information about any type of data.This functionality then links to the data sources, business applications, data lakes, data quality systems, and all sources of metadata to create a responsive system – essentially aligning the data to the business.These connections enable changes to be detected and policies immediately applied, without manual steps, driving active data governance.The operating model feature enables the business user to create analytics models with specific definitions and KPIs, and easily search for data while having a full view and understanding of the data, how it differs and where it comes from, and trust that information because the data is linked to the data governance platform. For example, when running analytics reports on customer loyalty, the data collected from website interaction and that of financial transactions on the backend can offer vital clues


70 percent of a

data analyst’s time is spent on

preparing and

analyzing data on questions

already answered.

40 | ORMS Today


February 2018

to customer buying behaviors, even when they have different meanings.This knowledge and understanding of the different but trusted data and meanings promotes data consumability and drives accurate insights and predictions. 3. Collaboration. After users have created their analytics models and prepared their data, they need the ability to explore the data in a way that suits their needs and objectives. First, they need to ensure algorithms are applied correctly and business rules are added, but also critical to the tools’ interaction is collaboration. A data catalog integrates the work of colleagues who may be looking for similar analyses, and can point the user to data sets they may have already created.This simplifies the process of finding relevant and comparable data and saves time if similar reports have already been developed. Because users can tag, document, and annotate data sets in the data catalog, the data is continuously enriched, increasing its value, eliminating data silos and encouraging collaboration and crowdsourcing. 4. Trust. Because policies and rules are applied to the data through governance and because data owners have been assigned and changes/updates are made consistently, users can feel confident knowing the data they’re using (including shared data) for analytics is data they are allowed to use, and/or has been previously approved, and falls under data privacy regulations.This is particularly essential when considering today’s increased regulatory environment and the nature of analytics, pulling data insights from different sources (including customer information) to apply them in external efforts. 5. Machine learning. By incorporating machine learning functionality via semantic search capabilities, the catalog can serve up increasingly relevant data to users over time and offer an automated and efficient way to improve data searches to be used in analysis.This is the Amazon-like feature mentioned earlier, “consumerizing” the user’s data discovery experience and analytical capabilities. As organizations adopt more self-service tools for BI and expand their analytical capabilities, leveraging a data catalog with these capabilities tied to data governance will give them confidence in knowing their business insights are based on trusted data.This is when we’ll start to see the true value of SSA tools in helping to drive business forward. ORMS Paul Brunet is vice president of product marketing at Collibra. An earlier version of this article appeared in Analytics magazine.


Editor-in-Chief, INFORMS Journal on Computing Deadline for Nominations: April 1, 2018 The final term of Professor David Woodruff as editor-in-chief of the journal INFORMS Journal on Computing will end on December 31, 2018. Members of INFORMS are invited to participate in this process to propose a new candidate. Qualifications for the editor-in-chief of INFORMS Journal on Computing and further information about the journal is available on the website: Please submit comments, recommendations, and nominations (including self-nominations) to the committee by April 1, 2018.



Biennial survey of vehicle routing software focuses on recent and future trends that are driving the industry forward. Image © alphaspirit |

By Peter Horner

Innovation powers dynamic VR sector Like a shark, vehicle routing (VR) software vendors have to keep moving forward or they’ll lose market share … or disappear from the marketplace altogether.Whether it’s responding to ever-increasing demands from customers or leveraging and integrating new technology in the big data era, the VR industry is constantly evolving and looking ahead. In such a dynamic and highly competitive environment, customers and clients benefit as higher expectations drive innovation. Further down the road lurk yet more potential industry disrupters such as driverless vehicles and delivery drones.We’ll get to those possibilities a little later, but first, let’s focus on the 2018 edition of OR/MS Today’s biennialVR software survey, particularly the trends and changes that have impacted the industry since our last such survey and the expected trends and changes that will likely make an impact in the foreseeable future as conveyed by respondents. Recent Trends The march toward cloud-based solutions or software-as-a service (SaaS), noted by Hall and Partyka in the 2016 survey [1], continues to accelerate. “Companies are more open to use SaaS and cloud-based technologies since they’re not new to them,” says Rubin Filter of Intelligent Routing Products when asked what was the biggest change he’s seen since the last survey. 42 | ORMS Today


February 2018

Adds Dr. Marc Gerlach, managing director of DNA Evolutions GmbH, “A major trend was the shift from in-house installations to web-based planning and SaaS solutions.” Meanwhile, the demand for same day or even hourly delivery service in some cases has created “greater need for real-time optimizers capable of same day re-routing on short notice to handle short delivery timescales – two-hour delivery for goods, 45 minutes for hot food,” says Dr. Philip Welch, founder of Open Door Logistics. Greg Wietholter, principal, Route Solutions, notes that mobile offerings have been bolstered as smartphones and tablets have become standard equipment for workers in the field, adding that the industry continues to move toward paperless, instant and fully synchronized operations. Filipe Carvalho, general manager of Widescope, offers a long list of recent trends, including predictive traffic while planning, embedded real-time traffic usage at execution time, the ability to optimize very large problems, the ability to use data to infer new rules and constraints in the future, and the adoption of mobile technology by drivers and recipients. Alessandro Lori, head of routing and data science at Fleetmatics, emphasizes the role artificial intelligence has played in recent years at his company. “Artificial intelligence and new data assets let us deliver to our customers a new product experience where

S route planning is based on customer needs and the latest information available from drivers and vehicles,” he says.

REFERENCE 1. Hall, R. and Partyka, J., 2016, “Higher expectations drive transformation,” OR/MS Today, Vol. 43, Nov. 1, pp. 40-47 (February 2016 issue).

What’s Ahead So what do the next few years hold for the VR software industry? Look for more regulatory changes, real-time routing requirements and still higher customer expectations and a continued move to cloud computing and mobile tools, along with highly automated planning and the consolidation of vendors through acquisition, say survey respondents. Widescope’s Carvalho anticipates optimization of larger parts of the supply chain simultaneously from the salesperson routes to the sourcing and warehousing to the long-haul transportation and final mile delivery. DNA Evolutions’ Gerlach expects to see more expert systems for specific functionalities. “Automatic tour planning and optimization will become more and more just one component next to others in integrated ERP solutions,” he says. Intelligent Routing’s Filter predicts client companies will look for the simplicity and scalability of cloud-based services technology, but they would like to stay in control of their data. “Self-hosting is one traditional option, but it does not scale well for smaller companies,” he says. “With services platforms like Amazon Web Services, the option of self-hosting is now extended to the additional option of full data control without the need of IT infrastructure. This allows for deployments where companies can stay in control of their data on a single platform while running the latest SaaS optimization solution.” To Infinity and Beyond According to the 1960s TV cartoon “The Jetsons” and numerous science fiction movies, we should all be driving flying cars. Five years ago, Amazon CEO Jeff Bezos imagined a fleet of drones delivering packages to businesses and homes.Today, driverless vehicles are plying the streets of America, albeit in very small numbers and largely prototype demonstration situations. Separating fantasy from reality for a moment, what will be the impact, if any, of drones and driverless vehicles on the VR industry looking further out into the future? Tim Pigden, managing direc-

tor and co-founder of Optrak Distribution Software, summed it up this way:“Drones are marginal (outside of small parcel deliveries). Driverless vehicles are dependent on adoption rate; the bigger consideration in the short term is ‘platooning’ and the associated additional algorithmic challenges. For multi-drop work, robotized deliveries are a long way off. Until then, personnel (drivers, crew) are still essential for completing the delivery process.” This Year’s Survey This year’s abbreviated survey includes responses from companies who responded by the deadline to an email and online questionnaire. The information provided (pages 38-39) is self-reported and unverified. Many other VR software vendors and products are out there, and we look forward to adding their names and products to the online version of the survey.Vendors who missed the deadline and would like to add their software product to the online version of the survey can do so by visiting: Hall and Partyka’s advice [1] in selecting a vehicle routing product holds true today as it did two years ago: “In selecting a vehicle routing product, look for vendors that have experience serving similar industries to your own, and test the software on a representative data set to assess the quality and speed of solutions. Ask for references and determine whether any prior customers have switched to another product and why. And look ahead to see whether the company has the capability to maintain and update the software to meet your future needs. Consider total cost of ownership, including license costs, staff support and future upgrades and maintenance.” As mentioned at the start, vehicle routing software companies need to be constantly moving forward or they will die.While the future is uncertain, it’s clear that new challenges and opportunities await … and high-end analytics and data-driven solutions are the keys to driving ultimate success in vehicle routing solutions. ORMS Peter Horner ( is the editor of OR/MS Today and Analytics magazines.

VENDOR DIRECTORY Clear Destination 389 - 4000 Saint Ambroise Montreal, Quebec H4C 2C7 Canada 514-933-8686

DNA Evolutions GmbH Marlene-Dietrich-Strasse 5 Neu-Ulm 89231 Germany +49 731 98588736

Fleetmatics 1100 Winter Street, Suite 4600 Waltham, MA 02451 866-844-2235

Intelligent Routing Products (Pty), Ltd. Centurion, Pretoria, Gauteng 157 South Africa +27 12 743 6789

MJC2 33 Wellington Business Park Crowthorne, RG45 6LS United Kingdom 0044 1344 760000

Open Door Logistics Unit 5 Tarlings Yard, Church Road, Bishops Cleeve, Cheltenham GL52 8RN United Kingdom

OptimoRoute, Inc.

Scientific Logistics

2657 Alma Street Palo Alto, CA 94306 +1-855-338-2838

Optrak Distribution Software, Ltd. Suite 6, The Maltings Hoe Lane, Ware, Hertfordshire NW6 1SE United Kingdom

75 Fifth Street Suite 363 Atlanta, GA 30308 678-388-9643

Wide Scope

RouteSolutions 3460 Marron Road Suite 103-137 Oceanside, CA 92056 858-541-2738

February 2018

Avenida da Repùblica 24 - 5º 1050-072 LISBOA Portugal +351 213156312


ORMS Today

| 43

16 GB

200 GB

Routing Functions

Computa with 50 tion Time to S o Hard-timRoutes, 1,000 lve Problem Sto e Windo ws (Spe ps, Two-Hour cify Platf orm) What Ty p in the S es of Algorithm oftware s (Open E are Employed nded)? Node R outing Arc Rou ting Same D ay Re-r outing Daily Ro uting Weekly Routing Route P lanning & Analy Ability to sis Create Te rritories Utilizes R e a ltim to Re-a ssign S e Traffic Inform tops Am Utilizes ong Dri ation R vers e a ltime Tra to Re-s ffic Info equenc rmation e Stops Provides fo r Drivers Dynamic Turn-byUtilizes Turn Inst H ructions Collecte istorical Travel d from M & obile De Stop Time vices

Hard Dis k




Process or Spee d

Recomm ended H ardware

Maximum Size of Problem Solvable by the System c. Numb er of Sim ultaneo us Term inals

Window s iOS Android Web-ba sed Soft ware as a Servic e (SaaS ) a. Num ber of S tops

Software Product Listing

Platforms Supported

b. Num ber of V ehicles

Year Introduced

ClearD Optima

2008 y y y y Unlmtd. Unlmtd. Unlmtd.


1990 y – – – Unlmtd. Unlmtd. Unlmtd. Depends Depends Depends Depends A few MJC2- y y y y y y y y y y y on on on on seconds, develappliappliappliappli- “stand- oped cation cation cation cation ard” in-house PC/ algorilaptop thms

Intelligent Routing

2012 – – y y


2006 y – – y Unlmtd. Unlmtd. Unlmtd.

ODL Studio

Min. Problem 2014 y – – – No set Unlmtd. Unlmtd. Modern limit but PC dual core depen<2000 2 GHZ dent, recommin. mended 4 GB


2013 – y y y Unlmtd. Unlmtd. Unlmtd.

Stand. N/A, N/A, N/A, Less PropriPC with webwebwebthan etary a web based based based 30 algoribrowser software software software seconds thms


2005 y – – y Unlmtd. Unlmtd. Unlmtd.

Intel I7 2 GHz + or equivalent


2013 – – – y Unlmtd. Unlmtd. Unlmtd.

Stand. PC with a web browser


Clear Destination


Intelligent Routing Products (Pty) LTD

DNA Evolutions GmbH

Open Door Logistics

OptimoRoute Inc.

Intel 1 GHz processor

Good internet connectivity

PC, 2 GHz Windows Server, Multicore

2 mins.

Heuristic y y y y y y y y y y y

5 mins.+ (acceptance setting dependent)

Multiple y – y y – y y – – y y Heuristics, dynamic select

4 GB

10 GB

3 mins.

Combi- y – y y y y y – – – – nation of construction, SA, GA

10 GB

1-5 mins.

Heur- y – – y – y y – – – – istics (Jsprit)

y – y y y y y – – y –

4 GB

100 MB

1 to 5 mins. SaaS

Propri- y – y y y y y – – – y etary metaheuristic

2010 y y y y Unlmtd. Unlmtd. Unlmtd. Depends on number of stops/ vehicles

2 Ghz

16 GB

100 GB

< 10 mins.

Meta- y – y y y y y y y y y heuristics, MIP, CP

Scientific Logistics Cloud-based Route Optimization

2011 – – – y Unlmtd. Unlmtd. Unlmtd. Browser UI




StreetSync Pro

2011 y y y y Unlmtd. Unlmtd. Unlmtd.

2 GB+

2 GB+

Optrak Distribution Software Ltd


Wide Scope

Scientific Logistics


44 | ORMS Today


February 2018

PC w/ 2 GHz+ HighSpeed Internet Access

5-10 Construc- y – y y y y y – – – y mins. tion heuristics and LNS

1+ Numer- y y y y y y y y y y y mins. ous depending on # of CPUs Several Propriminutes etary

y – y y y y y y y y y

Solution Algorithms


Part of a Suite that Provides:


Types of Fleets that Other Special Features Currently Use the Product


Most Significant Installations


Single S ite Lice nse (50 routes) a. License Fee Inclu de Map for One R egion? Source of Mapp ing Use d in You r Offerin g b. Insta llation S upport Cost ($/h our) c. Typic a Installa l Support Hou rs tion (50 routes) Needed for Conside ration o f Driver Geogra Skills & phic Re Needs? striction Driver H s? ours of S (e.g.,rest e rv ic e Rules breaks, 11 -hour rule , etc.)? Travel T imes fro m Extern al Sourc es? If yes, w hat is th e sourc e? Display Crowd-s ourced Informa tion? If yes, w hat type of inform ation? Are Ma p Stree t Views A RFID Sc vailable anner ? Supply Chain M a n a g Custom ement S er Orde oftware r Proce Compute ssing S ystem Fire or r-Aided Dispa Emerge tch for Police, ncy Veh icles Assigns Individu a l D Turn-by rivers to -Turn R Routes oute Ins Load M truction anifest s Loading Plan for Trucklo ETA Auto ad maticall y Sent to Local P the Cus ickup a tomer n d Delivery Long-h aul Les s than Tru Long-h ckload aul Truc kload Courier Buses Taxis Service Fleets Emerge ncy Serv ices (pol 24 by 7 ice, fire, Live Cu etc.) stomer Support Service Drivers Using S iOS martph ones? Android Window s Support Drivers Using Ta Utilize a blets? n App S tore for Mgmt. Distribu Use Dev tion? ices to Monitor Drivers ? Numbe r of Com panies Using S oftware

Pricing Information

Per stop y HERE pricing


y Compatible with most standard data

$65/ y OSM vehicle/ month call for bulk $20K

– Google Maps, Open Street Maps

No y Open license Street cost Map software is free $19$39/ driver/ month






y y y y


y y y y – y y y y y y y – y – – y – – y y y – y y y 101-500 Many big box retailers selling furniture appliances and mattresses

y – y y – y y y y y y y y y – – y – y y y y y y – – 501-1000 –

0 hrs. y y y y Android y Android y – – y – y y y – y y – – y – – y – – y – y – y y y 1-100 – with Users’ Users’ optional Choice Choice support (e.g. (e.g. Google) Waze) 7 hrs.

y y y y Google – Maps

– – y y – y – – – – y y y y – – y – – – – – – – – – 101-500 Retail logistics solutions by integration of our optimization components into ERP SW suites.

Contact 1-4 hrs. y y – –

– – – – – y – – – – y y y – y – – – – – – – – – – – 101-500 –

y Google Included None y y y – Maps in (SaaS and Open monthly Solution) Street subscripMaps tion

y – – – – y y y – – y y y y y y y – – y y y – y y y 101-500 –

y TomTom Global Maps

y – – – – y – y y y y y y y – – – – – y – – – y – y 1-100 –

y – – – – y y – – – y y y y y – y – y y y y y y y y 101-500 –

Please y All contact major Fleetvendors matics


$187.50 1- 2 hrs. y y y y Source – per come hour from mobile devices

y Open Street Maps


Depends y y y – on customization req.


16 hrs.

y y y –

y y y y TomTom, y Traffic, y y y y y y y y y y y y y y – y y y y y y y y y y y 101-500 Retailers, Basaardelivery maintenance and Soft, times support services, INRIX carriers

Contact y Various Contact Contact y y y y Various – premium premium suppliers

y – y – – y y – – y y y y y – – y – y y y y y – y y 1-100 Young’s Market

www. y HERE Included Typically y y y y Via y Via y y – y – – – – – – y y y y – – y – – y y y y y y y 101-500 Coca-Cola Route(formerly unnecTomTom TomTom Enterprises, Solutions. NAVTEQ) essary TeleTelemaGhirardelli Chocolate, com matics ticsMosquito Squad, lists inteintegraState of Michigan, prices gration tion City of San Francisco

February 2018


ORMS Today

| 45



ANALYTICS 2018 Baltimore, MD | April 15–17

If you are a leader in the Analytics profession, lead Analytics projects, or are just starting out in your career, you will gain valuable, real-world insights on successful analytics strategy. Join us in Baltimore to network with nearly 1,000 professionals from more than 400 companies within essentially all industries that can help your career and business.

Plenary Speakers: Bill Schmarzo, CTO, Big Data Practice Dell EMC Global Services Bruce D. Greenstein, CTO U.S Department of Health and Human Services (HHS)

Emerging Topic Tracks include: Revenue Management & Pricing

Analytics Process

Supply Chain Analytics

Decision & Risk Analysis

Analytics Leadership

Emerging Analytics

Analytics on Unstructured Data

Marketing Analytics

Analytics in the Public Sector


news Valuable content, contacts await analytics conference attendees Considering attending the 2018 INFORMS Conference on Business Analytics & Operations Research? Nowhere else in the world will you find the analytics and operations research expertise and experience across so many industries coming together for a threeday event. On April 15-17, a thousand professionals from more than 400 companies and essentially all industries will convene in Baltimore to hear 150 carefully chosen talks from the best in the business. Whether you are a leader in the analytics profession, lead analytics projects or are just starting out in your analytics profession, the conference offers valuable content. In addition, over the three days, the conference will provide plenty of planned networking opportunities with contacts who can help your career and your business. Professionals with tried and true experience will tell you how analytics have worked for them, the impact on the success of their businesses and how you can apply analytics more effectively. The INFORMS staff, Invited Track Committee and Selects Committee have thoughtfully handpicked industry giants and dynamic speakers for nine tracks each day created to give you useful techniques and practical information regarding all aspects of business analytics. Topic tracks include: analytics process, decision and risk analysis, revenue management and pricing, supply chain, marketing analytics, unstructured data, analytics leadership, emerging analytics and analytics in the public sector. The wide range of opportunities available make this conference a unique event for anyone looking for insight into helping decision-makers understand and utilize analytics to their highest potential.

INFORMS has always taken great care in its keynote speakers and this year is no exception. Leading off the conference is keynote speaker Bill Schmarzo, CTO Bill Schmarzo within Dell EMC’s 2,000-plus-person consulting organization and author of “Big Data: Understanding How Data Powers Big Business” and “Big Data MBA: Driving Business Strategies with Data Science.” Schmarzo works with organizations to identify where and how to start their big data journeys. With more than three decades of experience in data warehousing, BI and analytics, Schmarzo consistently shares his expertise as a teacher, author and speaker and was recently ranked as the No. 4 “Big Data Influencer” worldwide by Onalytica. Schmarzo holds an MBA from the University of Iowa and a bachelor of science degree in mathematics from Coe College. If you have followed Schmarzo on LinkedIn, you already know that he has provocative, interesting and impactful ideas, making him the perfect speaker to kick off this event. For more, visit: https://infocus. Bruce Greenstein, chief technology officer from the Department of Health and Human Services, will also deliver a keynote. Greenstein pro vides leadership and direction to ensure Bruce Greenstein that HHS effectively

Inside News 49

International Conference preview


Roundtable Report


Subdivision Awards


Simulation Society awards





Essential practice skills workshop Patrick Noonan, a former management consultant, business owner and college professor, will lead two-day workshops in Denver and the Washington, D.C., area later this year as part of INFORMS’ Professional Development and Continuing Education program. The workshops, entitled, “Essential Practice Skills for High-Impact Analytics Projects,” will be held June 20-21 at the University of Denver in Denver and Sept. 19-20 at the AMA Washington D.C. Area Executive Conference Center in Arlington, Va. In the workshops, Noonan combines a hands-on teaching and learning approach honed from his faculty positions at Harvard, Duke and most recently Emory University, with real-world experience from his business career starting at global management consulting giant McKinsey & Co. By the end of the course, participants will: 1) Learn to link their subject-matter expertise to the challenges of messy, unstructured problems, organizational noise and non-technical decisionmakers; and 2) Understand best-practice techniques, including: problem statement summaries, issue trees, interview guides, work plans, sensitivity analysis, stress-testing recommendations, the “Pyramid Principle” of story logic, story-boarding, slide-craft, delivering presentations and fielding Q&A. Participants in the workshop will learn and hone the skills they need to turn analytical insights into organizational actions that drive innovation, growth and efficiency. Past participants have come from a wide variety of industries, including manufacturing, healthcare, finance, transportation and many more, and have ranged from CFOs and data scientists to research analysts, business intelligence professionals, academic program directors. For more information or to register, visit: Essential-Practice-Skills-for-High-ImpactAnalytics-Projects. ORMS

Analytics Conference, continued on p. 48

February 2018


ORMS Today

| 47

new s Analytics Conference, continued from p. 47

uses data, technology and innovation to improve the lives of American people and the performance of the operating divisions across the department. Prior to his current appointment, Greenstein held a cabinet post with Louisiana Governor Bobby Jindal as the secretary of the Department of Health and Hospitals, where he led 11,000 employees with a budget of $9 billion. The Department of HHS has undertaken a huge analytics journey, which will affect health policy of all 330 million Americans. For more, visit: bruce-greenstein/index.html. As the general chair for this year’s conference, I am extremely honored to be working with experienced, successful and respected analytics professionals to help select key areas of interest and exciting speakers from these areas. Our knowledgeable volunteer members come from Deloitte, Accenture, Intel, Amazon, GM, American Airlines, Gartner, Kroger, Kromite, MITRE, academia and the government. To give you an idea of the level of our speakers’ analytics ability combined with excellent communications, a few of the speakers we have already confirmed include: Robert Moakler (Facebook), Anu Raman (Monsanto), Robert Phillips (Uber), Jim Guszcza (Deloitte), Mike Bentley (Revenue Analytics), Anuradha B h a m i d i p a t y ( I B M ) , G r e g Pa r n e l l (University of Arkansas) and Michael Feindt (Blue Yonder/CERN). In addition, the conference will offer an Analytics Career Fair, which brings dozens of top employers to Baltimore, all looking for topnotch analytics professionals. Starting on Sunday, April 15, the career will be open throughout the conference, so bring your resume and meet representatives from many companies who will be interviewing candidates all three days. One of the annual highlights of the conference, the 2018 Edelman Gala and presentation of the Franz Edelman Award for achievement in operations research, will take place on Monday evening, April 16. This truly international award has been presented since 1972 and has celebrated and honored outstanding applied work in business 48 | ORMS Today


February 2018

Since 1729, Baltimore has served as a vital hub of discovery and growth in Maryland. The city played a defining role in the founding of the United States, and has grown from a bustling port city to an industrial center to a community focused on service. Image © Jon Bilous |

and government from around the world. This year’s finalists include: • China National Petroleum Corporation (CNPC) • Europcar, and ACT Operations Research – OPTICAR system • Federal Communications Commission, which recently completed the world’s first twosided auction of valuable low-band electromagnetic spectrum • Intel, fully automated Multi-Echelon Inventory Optimization (MEIO) • Pediatric Heart Network and Georgia Institute of Technology, creators of clinical practice guidelines (CPG) • Turner Broadcasting System, Inc., TargetingNOW and AudienceNOW Nowhere else will you find such a collaborative, comprehensive approach to analytics and operations research practice than at the INFORMS Conference on Business Analytics & Operations Research. From talks involving real-world case studies to networking with industry giants, you will gain the blueprints you need to solve business problems in every sector. Join us April 15-17 in Baltimore at the beautiful

Baltimore Marriott Waterfront. The early registration rate of $1,175 for INFORMS members and $1,470 for non-members ends March 12. If you’re a newcomer, we encourage you to come check us out and enjoy a further discounted rate of $1,070. This is an exciting time for those of us in the analytics and operations research profession. Nearly every company and government organization is considering how to improve its use of data, analytics and operations research. Companies have learned that to ignore this area is to lose ground to the competition. Governments who ignore it risk inefficiency and ineffectiveness. The Edelman competition plainly shows how people and the environment are being helped throughout the world through better use of resources, more effective distribution of medicine and food, and overall better decision-making. We hope to see you in Baltimore. Learn. Connect. Enjoy! For more details and information or to register for the conference, visit the conference website: http://meetings2.informs. org/wordpress/analytics2018/. ORMS Jack Kloeber, principal, Kromite, LLC, is general chair of the 2018 INFORMS Conference on Analytics & Operations Research.

INFORMS International Conference set for Taipei on June 17-20 The 2018 INFORMS International Conference will be held in Taipei, Taiwan, on June 17-20. The theme of the conference is “A Better World through O.R., Analytics and A.I.” The conference will showcase advancing trends in deep learning and their impact on business and society. Anyone interested in learning the latest technological advancements and industry trends from a global line-up of experts should attend. With more than 400 speakers and attracting about a thousand researchers and practitioners, the INFORMS International Conference is one of the largest international conferences in operations research and management science. Tracks will span the full range of OR/MS specialties, including, but not limited to, financial technology, education, healthcare, maritime operations, retail services, sports and transportation. Through a series of widely diverse topics, the conference will help attendees discover advances in deep learning and artificial intelligence.

Program highlights include four plenary and keynote presentations from Christopher Tang, UCLA Anderson School; Richard C. Larson, MIT; David D. Yao, Columbia University; and Dr. Shmuel S. Oren, UC Berkeley. In addition to the technical tracks, plenaries and keynote presentations, the Taipei is the political, economic, educational and cultural program also includes two recenter of Taiwan. Image © Chih Hsien Hang | ceptions primarily focused on networking with international colleagues. Registration fees for this event start at The conference will take place at the $300 (U.S.) for students and retired attendees Grand Hyatt Taipei in the Xinyi District of and $630 (U.S.) for INFORMS members. Taipei. Taipei City is the political, economic, If you are interested in submitting a educational and cultural center of Taiwan presentation or attending the conference, Island. In addition to being a major hub of visit the conference website (http://meetGreater China and part of a major high-tech for industrial area, Taipei City is known for its many more information. ORMS night markets featuring individual stalls selling a – Jessica Bennett, meeting planner, mixture of food, clothing and consumer goods. INFORMS

2018 ANALYTICS CAREER FAIR Baltimore, Maryland | April 15–17

Find the Right Analytics Professional to Make Sense of Your Data. • • • •

Find seasoned professionals you need – over 800 analytics professionals expected Provide your recruitment materials in a casual setting Arrange on-site meetings in private booths Enhance your visibility with an ad in Analytics or OR/MS Today

QUESTIONS? Email us at, call (800) 4-INFORMs, or visit our website at

February 2018


ORMS Today

| 49

new s Recap: Fall 2017 Roundtable meeting in Houston The fall 2017 Roundtable meeting was held Oct 21-22 at the Hilton Americas Hotel in Houston with a meeting theme of “Academic-Industry Partnerships.” To help facilitate networking opportunities valued by Roundtable members, a guided tour of the National Aeronautics and Space Administration (NASA) was held on Saturday afternoon followed by a reception. No Roundtable member company can be successful without the talent pool that it hires. With the fast-paced change of the analytics field, Roundtable members wanted to hear what INFORMS and some academic institutions are doing to address the challenge of maintaining the analytics talent pool and the evolution of the academic curriculum to focus on valued skill sets (technical and soft skills) that make a successful employee. The day was broken into four sections: 1) examples of successful academic-industry partnerships, 2) the INFORMS Masters in Analytics Program Directors (MAPD) program and changes in curriculum to address practitioners needs, 3) introduction to the Ford Academic Alliance program, and 4) new hire expectations from their jobs. The day was kicked off by Ravi Ahuja, president and CEO of Optym, who talked about his experiences in consulting for several railroads, trucking and mining companies. According to Ahuja, industry has a mixed experience with optimization or black box solutions. He advocated the need to focus on interactivity and graphical solutions relying more on heuristics. He also emphasized that successful O.R. practitioners need not just strong math skills, but also strong programming skills. Academics need to teach these skills to their students. He also called on practitioners to provide sabbatical opportunities to faculty and provide students with real data and internship opportunities. Martin Savelsberg, the James C. Edenfield chair and professor at the H. Milton Stewart School of Industrial & Systems Engineering, Georgia Tech, followed with a talk about various partnership arrangements with practitioners such as contract research, private consulting and research grants. The factors 50 | ORMS Today


February 2018

that need to be balanced in these engagements, he said, are the needs of students with the needs of the company, including intellectual property and the transfer of code and knowledge. He spoke about his experiences with DayJet, Saia, UPS and Grub Hub. Each project was structured differently Houston hosted the Roundtable meeting in conjunction with the in terms of intellectual 2017 INFORMS Annual Meeting. Source: Thinkstock property and the roles of his team (idea generator versus responsibility within 60 days of graduation. The program for the code base). emphasizes technical skills, business acumen After a short break, the Roundtable and impactful professionalism. members moved to the next section of the Camm’s talk was followed by a “What day – learning what academic institutions are does the Roundtable think?” session in which doing in response to the shortage of analytics the Roundtable members were divided professionals and changes in the technical into two groups to focus on academics and landscape. Melissa Bowers (associate practitioner strengths and opportunities. professor, University of Tennessee) and Each group was assigned one of two roles Jeff Camm (associate dean, Wake Forest – academics or industry – and each group University) tag-teamed in this section. had both academics and practitioners. The In the past seven years Bower’s survey goal of each group was to identify a couple results have shown that masters in analyt- of strengths in how academic institutions ics programs have grown from a handful to are working currently along with a couple nearly 200 in the U.S. alone. These programs of opportunities to improve. The same have strong industry partnerships and industry exercise was repeated for industry. What advisory board members, and they focus on are practitioners doing well and what can be soft skills and hands-on experience. Each of improved? these programs have a requisite capstone inIn the third section of the day, Tim ternship requirement that teaches students to Maull, University Programs manager from create insightful results and communicate to the Ford Motor Company, talked about various leadership levels. Bowers also talked Ford’s industry alliance program that proabout an INFORMS-sponsored program, vides a framework for different teams and the Masters in Analytics Program Directors organizations within Ford to collaborate (MAPD) meeting, that was attended by more with universities. than 40 program directors to discuss the After lunch, three newly hired analytics common challenges. professionals (Sushil Raj Podel, UPS; Nathaniel Jeff Camm, who has had several decades Richmond, BNSF Railway; and Hadi of successful experience with industry Panahi, Monsanto) talked briefly about their consulting as a University of Cincinnati faculty expectations at their respective companies. member, talked about his new master’s in Their talks provided insights on some simple analytics program at Wake Forest University policies and conditions that provide meaning that is focused on three principles: the voice and energy to young employees. of the customer, program benchmarking and The Roundtable meeting concluded faculty subject matter experts. Based on these with a short business meeting conducted principles, Camm designed an 11-month by Kathy Lange. ORMS program; first set of 37 graduates were hired – Pooja Dewan

2017 INFORMS Subdivision Awards The following awards were presented by the respective INFORMS societies, sections, interest groups, forums, fora, etc., at the 2017 INFORMS Annual Meeting in Houston, in October. ANALYTICS SOCIETY Innovative Applications in Analytics Award


COMPUTING INFORMS Computing Society Prize

Best Working Paper Award

First-place recipients: Wes Chaar, Jose Antonio Carbajal, Peter Williams Recognized work:

“Audience Targeting Solutions Powered by Advanced Analytics” Second-place recipients: Valentina Ferretti, Alessandro Degioanni, Maurizio Arnone Recognized work:

“How to Regenerate Disused Railways? An Integrated Analytics Approach”

APPLIED PROBABILITY Best Publication Award Recipients: Anton Braverman, Jim Dai, Jiekun Feng, Itai Gurvich

Best Publication Award Recipients: Siva Theja Maguluri, Rayadurgam Srikant

Best Student Paper

Matt Saltzman (far left) congratulates Juan Pablo Vielma and Shabbir Ahmed (second and third from left). Ruomeng Cui and Dennis J. Zhang (second and third from left) receive Best Working Paper Award.

First-place recipients: Ruomeng Cui, Jun Li, Dennis J. Zhang Recognized work:

Recipients: Shabbir Ahmed, George Nemhauser, Juan Pablo Vielma

ICS Student Paper Prize

“Discrimination with Incomplete Information in The Sharing Economy: Evidence from Field Experiments on Airbnb” Second-place recipients: Bhavani Shanker Upari, Sameer Hasija Recognized work:

“Modeling News Vendor Behavior: A Prospect Theory Approach” Honorable mention: Eric Park, Yichuan Ding, Mahesh Nagarajan, Eric Grafstein Recognized work:

“Patient Prioritization in Emergency Department Triage Systems: An Empirical Study of Canadian Triage”

Berk Ustun (center) receives ICS Student Paper Prize.

First-place recipient: Berk Ustun Recognized Work:

CONFERENCE ON SYSTEMS AND TECHNOLOGY Best Conference Paper Guido R. Lagos (center) receives Best Student Paper award.

Recipient: Guido R. Lagos, Georgia Tech Recognized work:

“On the Euler Discretization Error of Brownian Motion about Random Time.”


First-place recipients: Manmohan Aseri, Milind Dawande, Ganesh Janakiraman, Vijay Mookerjee Recognized work:

“Block Ad Block Users? Publishers’ Strategies and Impact on Revenue” Second-place recipients: Dezhi Yin, Samuel Bond, Han Zhang Recognized work:

“Are Helpful Reviews Persuasive? Effects of Anger in Online Word of Mouth”

“Learning Optimized Risk Scores from Large-Scale Datasets” Second-place recipients: Rui Gao, Christian Kroer, Hamed Rahimian

DATA MINING Best Paper Competition Recipients: Durai Sundarmoorthi, Chen Zhang, Rui Gao, Yixin Wang


Best Student Paper First-place recipients: Jiyong Park, Kunsoo Han, Byungtae Recognized work:

Heng Chen (right) receives Dissertation Prize

Recipient: Heng Chen

Recognized work:

“Service Improvement and Cost Reduction for Airlines: Optimal Policies for Managing Arrival and Departure Operations”

“Is Cloud Computing Green? An Empirical Analysis of Information Technology Outsourcing and Energy Efficiency” Second-place recipients: Nikhil Malik, Param Vir Singh, Dokyun Lee, Kannan Srinivasan Recognized work:

“When Does Beauty Pay? A Large-Scale Image Based Appearance Analysis on Career Transitions”

Valentina Ferretti (left) and Gabriella Csányi (right) receive Practice Award.

Recipients: Valentina Ferretti, Gabriella Csányi Recognized work:

“How to improve educational programs for underprivileged children? The impacts of valuefocused decision analysis”

February 2018


ORMS Today

| 51

new s Council Service Recipients: Debarun Bhattacharjya, Joe Hahn, Melissa A. Kenney, Yael Grushka-Cockayne, Andrea Cadenbach, Casey Lichtendahl, John Butler


ENRE Young Researcher

Best Publication Award in Energy

Student Paper Award

Adrda Yenipazarli (third from left) accepts ENRE Young Researcher award. Andy Sun (second from left) accepts Best Publication Award in Energy.

Bhavani Shanker Uppari (center) receives Student Paper Award.

Recipient: Bhavani Shanker Uppari Recognized work:

“Modeling News Vendor Behavior: A Prospect Theory Approach”

Recipients: Alvaro Lorca, X. Andy Sun, Eugene Litvinov, Tongxin Zheng Recognized work:

“Multistage Adaptive Robust Optimization for the Unit Commitment Problem”

Recipient: Adrda Yenipazarli

Recognized work:

“Managing new and remanufactured products to mitigate environmental damage under emissions regulation”

Student Best Paper Award

Best Publication Award in Environment & Sustainability

Publication Award Recipients: Neil Stewart, Stain Reimers, Adam J.L. Harris

Abdual Quddus (second from left) earned the Student Best Paper Award.

Frank P. Ramsey Medal Erin Baker and Senay Solak (third and fourth from left) receive Best Publication in Environment & Sustainability Award.

Recipients: Erin Baker, Senay Solak Recognized work:

Simon French (left) receives Ramsey Medal.

Recipient: Simon French Recognition:

Distinguished contributors in decision analysis

“Management of Energy Technology for Sustainability: How to Fund Energy Technology Research and Development”

Recipient: Abdual Quddus

Recognized work:

“Optimizing Electric Vehicle Charging Station Expansion with an Integration of Renewable Energy and Vehicle-to-Grid Sources”

FINANCIAL SERVICES Student Paper Competition

Best Publication Award in Natural Resources

E-BUSINESS Best Paper Award First-place recipient: Yan Huang, Stefanus Jasin, Puneet Manchanda

Student Paper Competition winner Allen Cheng.

First-place recipient: Allen Cheng

Recognized work:

“Level Up: Leveraging Skill and Engagement to Maximize Player Retention in Online Video Games” Second-place recipients: Ranmath K. Chaellappa, Rajiv Mukherjee, Zhan (Michael) Shi, T.S. Raghu, Zhongju (John) Zhang

Section Service Award Recipient: Hemant Bhargava

52 | ORMS Today

Recognized work:

Liying Mu (middle) accepts Best Publication Award in Natural Resources.

Recipients: Liying Mu, Milind Dawande, Xianjun Geng, Vijay Mookerjee Recognized work:

“Milking the Quality Test: Improving the Milk Supply Chain under Competing Collection Intermediates”


February 2018

“Clearinghouse Default Waterfalls: Risk-sharing, Incentives, and Systemic Risk” Second-place recipient: Yi Zhang Recognized work:

“Default Risk Premia and a Non-Linear Asset Pricing Model”

GROUP DECISION AND NEGOTIATION Best Paper Award First-place recipient: Igor Pyrko Second-place recipient: Rudolf Vetschera

Young Researcher Award First-place recipient: Tobias Langenegger Second-place recipient: Francis Marleau

HEALTH APPLICATIONS Pierskalla Best Paper Award

Nunamaker-Chen Dissertation Award First-place recipient: Yi-Jen Ian Ho Recognized work:

“The Impact of Social and Spatial Proximity on Consumer Choice in Digital Markets” Second-place recipient: Tianshu Sun Recognized work:

“Engineering Digital Sharing Platforms to Create Social Contagion: Evidence from Three Large Scales” Third-place recipient: Zike Cao

“Optimal Timing for Drug Sensitivity Testing for Patients on First-line Tuberculosis Treatment” Second-place recipients: Wenqi Hu, Carri W. Chan, Jose R. Zubizarreta, Gabriel J. Escobar, Hrayer Aprahamian, Douglas R. Bish, Ebru K. Bish Recognized work:

“An examination of early transfers to the ICU based on Physiologic Risk Score”

INFORMATION SYSTEMS SOCIETY Distinguished Fellow Award Recipients: Hemant Bhargava, Gedas Adamovicius

2016 Management Science Best Paper Award

Charles ReVelle Award

INFORMS ISS Sandy Slaughter Early Career Award

Moving Spirit Award

Recognized work:

Lifetime Achievement Recipient: Morton O’Kelly

“Three Essays on Economics of Online Communities and Platforms”


First-place recipients: Sze-chuan Suen, Margaret L. Brandeau, Jeremy D. Goldhaber-Fiebert


Recognized work:

Recipient: Gordan Burth, Mingfeng Lin, Marius Florin Niculescu, Sam Ransbotham

Sze-chuan Suen (second from right) and Margaret Brandeau (far right) receive Pierskalla Best Paper Award.

Third-place recipients: Yiwei Chen, Cong Si Honorable mention: Ruiwei Jiang, Johanna Mathieu, Shipra Agrawal

Recipients: Dan Steffy, Jay Weber

Judith Liebman Award Recipients: Andres Patricio Garcia Arce, Mohammad MoshrefJavadi, Rozhin Doroudi

Student Chapter Annual Awards Recipients (Summa Cum Laude): University of Texas at Dallas, University of South Florida Recipients (Magna Cum Laude): University of Michigan, Carnegie Mellon University, University of Toronto, University of Texas at Austin, University of Massachusetts Recipients (Cum Laude): Auburn University, Northwestern University, Koç University, University at Buffalo, Lehigh University, University of Florida, Virginia Tech, Columbia University, Northeastern University, Texas A&M University, Concordia University, Purdue University


Recipients: Prasanna Tambre, Lorin M. Hitt Recognized work:

Sibel Alumur Alev (right) receives award.

Recipient: Sibel Alumur Alev

MILITARY APPLICATIONS 2016 Koopman Prize Recipient: Jun Zhang

Recognized work:

“Introducing Terrorist Archetypes”

J. Steinhardt Prize Recipient: Richard Decko Recognition:

Outstanding contributions to military operations research.

MINORITY ISSUES FORUM Early Career Award First-place recipient: Krystal K. Castillo Honorable mention: Shadi Sharif Azadeh

Minority Issues Forum Paper Competition First-place recipients: Michelle Alvarado, Lewis Ntaimo Honorable mention: Laquanda Leaven, Xiuli Qu

“Job hopping, information technology spillovers, and productivity growth”

2017 Management Science Best Paper Award Recipients: Young-Jin Lee, Kartik Hosanagar and Yong Tan Recognized work:

“Do I follow my friends or the crowd? Information cascades in online movie ratings”

Krishnamurthy Iyer, Marian Epelman and David Lingenbrink (l-r).

First-place recipient: Krishnamurthy Iyer Second-place recipient: Velibor Misic

February 2018


ORMS Today

| 53

new s Minority Issues Poster Competition

Poster Competition winner Gian-Gabriel P. Garcia (front center).

First-place recipient: Gian-Gabriel P. Garcia Honorable mention: Toyya A. Pujol, Donald Richardson, Lauren N. Steimle

Best OM Paper in Management Science

Ming Hu, Guillermo Gallego and MSOM President Gad Allon (l-r).

Recipients: Guillermo Gallego, Ming Hu Recognized work:

“Dynamic Pricing of Perishable Assets Under Competition”

MSOM Biennial Practice-Based Paper Competition for the M&SOM Journal First place recipients: Hongmin Li, Scott Webster, Nicholas Mason, Karl Kampf, Turgay Ayer, Can Zhang, Chenxi Zeng, Chelsea C. White III, V. Roshan Joseph Honorable mention: Pavithra Harsha, Shivaram Subramanian, Joline Uichanco

Distinguished Service Award

Practice-Based Paper Competition First-place recipients: Turgay Ayer, Can Zhang, Chenxi Zeng, Chelsea C. White III, V. Roshan Joseph, Hongmin Li, Scott Webster, Nicholas Mason, Karl Kempf Honorable mention: Pavithra Harsha, Shivaram Subramanian, Joline Uichanco

Best Student Paper

Can Zhang of Georgia Tech (third from left) receives Best Student Paper Award.

First place recipient: Can Zhang Recognized work:

“Truthful Mechanisms for Medical Surplus Product Allocation” Second-place recipient: Jiankun Sun Recognized work:

“Robust Dual Sourcing Inventory Management: Optimality of Capped Dual Index Policies”

OPTIMIZATION Distinguished Service Prize Recipient: Pietro Belotti

Young Scholar Prize

Khachiyan Prize

Recipient: Mark Ferguson

Distinguished Fellows Award

Omar Besbes (left) and MSOM President Gad Allon.

Recipients: Omar Besbes, Enno Siemsen MSOM President Gad Allon (left) congratulates Jeannette Song.

Service Management Best Paper

Committee Chair Gerald Brown (left) congratulates Robert J. Vanderbei (center).

Recipient: Robert J. Vanderbei Recognition:

Lifetime achievements in the field of optimization.

Recipients: Jeannette Song, L. Beril Toktay, Sridhar R. Tayur

Farkas Prize

Best Paper Award

Peng Shi (left) and MSOM President Gad Allon.

Recipients: Itai Ashlagi, Peng Shi Recognized work:

Jason Acimovic (left) and MSOM President Gad Allon.

Recipients: Jason Acimovic, Stephen C. Graves

“Optimal Allocation Without Money: An Engineering Approach”

Kim-Chuan Toh (center) accepts the Farkas Prize.

Recipient: Kim-Chuan Toh Recognition:

Outstanding contributions to the field of optimization.

Recognized work:

“Making Better Fulfillment Decisions on the Fly in an Online Retail Environment”

54 | ORMS Today


February 2018

Student Paper Prize

Second-place recipients: Selin Merdan, Christine Barnett, Brian T. Denton, James E. Montie, David C. Miller

Railway Applications Student Paper Award

Recognized work:

“Data Analytics for Optimal Detection of Metastatic Prostate Cancer”

QUALITY, STATISTICS & RELIABILITY Best Refereed Paper Award Frans de Ruiter (right) accepts the Student Paper Prize.

First-place recipient: Frans J.C.T. de Ruiter Honorable mention: Will Ma

Recipients: Chenang Liu, Zhenyu (James) Kong Recognized work:

“An Integrated Manifold Learning Approach for Online Process Monitoring of Additive Manufacturing Processes”

Former Chair

Prize for Young Researchers Recipient: Hui Yang

Best Student Paper Award Recognized work:

Recipients: Alberto Del Pia, Aida Khajavirad

ORGANIZATION SCIENCE Best Dissertation Proposal Competition First-place recipient: Pedro Aceves

First-place recipient: Thomas Breugem Recognized work:

Recipient: Raed Al Kontar

Aida Khajavirad (second from left) and Alberto Del Pia (third from left).

Thomas Breugem (right) receives Railway Applications Student Paper Award.

“Nonparametric Modeling and Prognosis of Condition Monitoring Signals: A Transfer Learning Approach Based on Multivariate Gaussian Convolution Processes”

Best Student Poster Award Recipient: Ahmed Aziz Ezzat

“An Optimization Framework for Fairness-oriented Crew Rostering” Second-place recipient: Sofie Van Thielen Recognized work:

“Benefits of a Dynamic Impact Zone when Dealing with Train Conflicts” Third-place recipient: Pan Chang Recognized work:

“Equity-Oriented Ski-Stopping Optimization in an oversaturated urban rail transit network”

Problem-Solving Competition

RAILWAY APPLICATIONS Railway Applications Distinguished Member Award

Recognized work:

“The Linguistics Relativity of Collective Cognition and Group Performance” Second-place recipient: Evelyn Zhang

Tai-Chia Huang (second from left) won Problem-Solving Competition.

Recognized work:

“Intra-Organizational Mobility: Movers, Incumbents, and Communication Networks”

Recipient: Tai-Chia Huang

PUBLIC SECTOR O.R. Best Paper Competition Pooja Dewan (left) accepts Railway Applications Distinguished Member Award.

Recipient: Pooja Dewan

RAS Service Award Recipients: Xuesong Zhou and April Kuo Best Paper winner Ronghuo Zheng (second from left).

First-place recipients: Ronghuo Zheng, Tinglong Dai, Katia Sycara Recognized work:

“Jumping the Line, Charitably: Analysis and Remedy Donor Priority Rule”

REVENUE MANAGEMENT AND PRICING Section Award Recipients: Richard Ratliff, Garrett van Ryzin, Gustavo Vulcano Recognized work:

“Estimating Primary Demand for Substitutable Products from Sales Transaction Data”

Dissertation Award First-place recipient: Negin Golrezaei Recognized work:

“Efficient Policies and Mechanisms for Online Platforms” Honorable mention: Jake Feldman Recognized work:

“New perspectives on incorporating customer choice into Revenue Management”

February 2018


ORMS Today

| 55

new s Practice Award Recipients: Markus Ettl, Pavithra Harsha, Shivaram Subramanian, Joline Uichanco Recognized work:

“Omni-channel Markdown Optimization”


Best Paper in Transportation Science & Logistics

Distinguished Speaker Award

SERVICE SCIENCE Best Article Award First-place recipients: Steve Vargo, Kaisa Koskela-Huotari, Jaakko Siltaloppi

Alexandre Jacquillat (center) receives Best Paper in TSL Award.

Recognized work:

“Institutional Complexity as a Driver for Innovation in Service Ecosystems” Second-place recipients: Cong Feng, K. Sivakumar Recognized work:

“The Role of Collaboration in Service Innovation across Manufacturing and Service Sectors”

Best Cluster Paper Award First-place recipients: Anna Papush, Georgia Perakis, Pavithra Harsha Recognized work:

“A Data-Driven Approach to Personalized Bundle Pricing and Recommendation” Honorable mention: Guihua Wang, Jun Li, and Wallace J. Hopp, Renyu (Philip) Zhang and Maxime C. Cohen

IBM Best Student Paper Award First-place recipient: Wenchang Zhang Recognized work:

“Managing Market Thickness in Online B2B Markets” Second-place recipient: Hossein Jahandideh Recognized work:

“Dynamic Network Revenue Management for a Cloud Provider Hosting Interactive Applications” Third-place recipient: Seung-Yup Lee

John Paul MacDuffie (middle) receives Distinguished Speaker Award.

Recipient: John Paul MacDuffie

Best Dissertation Award (sponsored by the Lazaridis Institute)

Recipients: Alexandre Jacquillat, Amedeo R. Odoni Recognized work:

“An Integrated Scheduling and Operations Approach to Airport Congestion Mitigation”

Dissertation Prize

First-place recipient: Tugce Martagan Second-place recipient: Tian Heong Chan Honorable mention: Bo Cowgill, Hyunwoo Park, Douglas Hannah

Best Paper Award First-place recipient: Kevin J. Boudreau Second-place recipients: Prasanna Tambre, Lorin M. Hitt, Erik Brynjolfsson, Marco Tortoriello, Ray Reagans, Bill McEvily

TRANSPORTATION SCIENCE & LOGISTICS Robert Herman Lifetime Achievement Award

Recognized work:

Mehdi Behroozi (center) receives Dissertation Prize.

Recipient: Mehdi Behroozi

Recognized work:

“Robust Solutions for Geographic Resource Allocation Problems” Honorable mention: Marlin Ulmer, Jelmer Pier van der Gaast

Outstanding Paper in Air Transportation

“Proactive Inpatient Bed Reservations from ED to Reduce”

SOCIAL MEDIA Best Student Paper Competition

TSL President Karen Smilowitz (left) and Moshe Ben-Akiva.

Recipient: Moshe Ben-Akiva

President’s Service Award Parshuram Hotkar (right), winner of Social Media Student Paper Competition.

Chiwei Yan (left) and TSL President Karen Smilowitz.

Recipients: Chiwei Yan, Jerry Kung Recognized work:

“Robust Aircraft Routing”

First-place recipient: Parshuram Hotkar Second-place recipient: Zhenhuan Sui

Andres L. Medaglia (third from left) receives President’s Service Award.

Recipient: Andres L. Medaglia

56 | ORMS Today


February 2018

Outstanding Paper in Freight Transportation and Logistics

INFORMS Simulation Society presents awards at WSC The INFORMS Simulation Society presented the following awards at the 2017 Winter Simulation Conference held in Las Vegas in December:

Jannik Matuschke (left) and TSL President Karen Smilowitz.

Recipients: Tobias Hanks, Felix G. Konig, Jannik Matuschke, Alexander T. Richter, Jens Schulz Recognized work:

“An Integrated Approach to Tactical Transportation Planning in Logistics Networks”

WOMEN IN OR/MS Advancement of Women in OR/MS Recipient: Ruth Williams

Lifetime Professional Achievement Award

Lee Schruben (left) receives award from Russell C. H. Cheng.

Ilya Ryzhov (left) receives award from Barry Nelson.

Doctoral Colloquium Student Sponsorship Award

Kathryn Smith (center) receives Doctoral Colloquium Student Sponsorship Award.

Recipient: Kathryn Smith

Monsanto Travel Award

Monsanto Travel Award recipients Ying Zhang and Shadi Hassani Goodarzi (third and fourth from left).

Recipients: Ying Zhang, Shadi Hassani Goodarzi

Family Care Travel Award

WORMS Family Care Travel Award recipients.

Recipients: Zhijie Dong, Yueran Zhuo, Yaarit Even, Unit Deniz Tursan, Irem Sengul Orgut

Outstanding Simulation Publication Award

Distinguished Service Award John Fowler (left) receives award from Enver Yucesan.

Winter Simulation Conference Best Student Paper Award

Michael Pearce (left) receives award from Emily Lada.

Winter Simulation Conference Diversity Award Wei Xie congratulates Vahab Vahdat, Rana Azghandi and Ange Lionel Toba (l-r).

The Lifetime Professional Achievement Award, the highest honor given by the Society, was presented to Lee Schruben, professor at the University of California, Berkeley. The award recognizes major contributions to the field of simulation that are sustained over a professional career, with the critical consideration being the total impact of those contributions on computer simulation. The Distinguished Service Award was presented to John Fowler, professor at Arizona State, in recognition of his longstanding, exceptional service to the simulation community. The award is for sustained service to the simulation community over at least 15 years that was acquitted with distinction in the areas of elected offices in simulation societies. The Outstanding Simulation Publication Award was presented

to Ilya Ryzhov, an associate professor at the University of Maryland, for the paper, “On the convergence rates of expected improvement method,” published in Operations Research. The award recognizes outstanding contributions to the simulation literature. The Winter Simulation Conference Best Student Paper Award was presented to Michael Pearce of the University of Warwick for the paper, “Bayesian Simulation optimization with input uncertainty.” The Winter Simulation Conference Diversity Award was presented to grad students Vahab Vahdat, Rana Azghandi and Ange Lionel Toba. The award aims to improve outreach and diversity among young researchers in the field of simulation while assisting eligible grad students to attend the Winter Simulation Conference. ORMS February 2018


ORMS Today

| 57

new s Meetings

People Brenda Dietrich, a Fellow and past president of INFORMS, has joined Cornell University as the Arthur and Helen Geoffrion Professor of Practice in the Brenda Dietrich School of Operations Research following a 33-year career at IBM. At IBM, Dietrich and teams she managed applied data and computation to business decision processes throughout IBM and for IBM clients. For over a decade Dietrich led the Mathematical Sciences function in the IBM Research division where she was responsible for both basic research on computational mathematics and for the development of novel applications of mathematics for both IBM and its clients. In addition, she has been the chief technology officer and strategist for IBM’s Business Analytics group, and led Emerging Technologies in the IBM Watson Group. Dietrich was appointed IBM Fellow in 2007 and vice president in 2008. She holds several honors from INFORMS including the Kimball Medal for service (2011), Distinguished Lecturer (2012) and the Impact Prize (2014). John D. C. Little, a pillar of the operations research community for more than 60 years and a member of the first class of INFORMS Fellows, celebrated John D. C. Little his 90th birthday on Feb. 1. An Institute Professor Emeritus at MIT’s Sloan School, Professor Little is perhaps best known for his proof of the queueing formula (l = lw), commonly referred to as Little’s Law. He was a founder of the field of marketing science. Little served as president of both the Operations Research Society of America (ORSA) and The Institute of Management Sciences (TIMS), which merged in 1995 to create INFORMS, after which he became the first president of INFORMS. Studying with O.R. pioneer Philip Morse at MIT, Little earned the first Ph.D. in operations research in 1955. 58 | ORMS Today


February 2018

INFORMS Annual & International Meetings 2018 April 15-17 INFORMS Conference on Business Analytics & Operations Research Marriott Waterfront Hotel, Baltimore Chair: Jack Kloeber, Kromite, LLC

June 17-20 INFORMS International Conference

Taipei International Convention Center Taipei, Taiwan Chair: Grace Lin, Asia University

Nov. 4-7 INFORMS Annual Meeting

Phoenix Convention Center & Sheraton Phoenix Hotel Phoenix Chair: Young-Jun Son, University of Arizona


INFORMS Community Meetings 2018 March 1-3 2018 INFORMS Organization Science Conference Park City Marriott, Park City, Utah

March 23-25 INFORMS Optimization Society Conference

University of Colorado, Denver Chair: Stephen C. Billups, University of Colorado-Denver

May 23-25 14th INFORMS Telecommunications Conference University of Hamburg, Hamburg, Germany Chair: Stefan Voss konferenzen/informs2018.html

June 14-16 INFORMS Marketing Science Conference Temple University, Philadelphia Chair: Xueming Luo, Temple University

April 14-16 INFORMS Conference on Business Analytics & Operations Research

June 21-22 2018 INFORMS Revenue Management and Pricing Section Conference

JW Marriott Austin Austin, Texas

Scotiabank Conference Centre, Toronto, Canada

June 9-12 INFORMS International Conference

Dec. 9-12 Winter Simulation Conference

JW Marriott & CasaMagna Marriott Cancun Cancun, Mexico

The Swedish Exhibition & Congress Center, Gothenburg, Sweden Chair: Bjorn Johansson, Chalmers University of Technology

Oct. 20-23 INFORMS Annual Meeting

Washington State Convention Center & Sheraton Seattle Hotel Seattle, Wash.

Go to for a searchable INFORMS Conference Calendar.

Elected to the National Academy of Engineering, Little received the Kimball Medal from ORSA and the Distinguished Service Medal from TIMS. Richard C. Larson, a past president of INFORMS, an INFORMS Founding Fe l l o w a n d t h e Mitsui Professor of Data, Systems and Society at MIT, was Richard C. Larson awarded the firstever Daniel Berg Lifetime Achievement Medal by the International Academy of Information Technology and Quantitative Management (IAITQM). The presentation was made at a recent IAITQM conference in New Delhi,

India, where Professor Larson delivered a keynote address. IAITQM created the medal in Daniel Berg’s name to honor his distinguished achievements in technology and service systems. The award citation noted that Larson has made significant contributions in both public and private sector services. His public-sector work includes urban emergency services, public health, technology-enabled education, postal services, urban sanitation, criminal justice and disaster management. His private sector research includes industrial gas distribution, home energy management and quality of customer service. An expert on queues and widely known as “Dr. Queue,” he is a past president of ORSA and a member of the U.S. National Academy of Engineering. ORMS


SPECIAL ADVERTISING SECTION | View Classifieds Online at:

The Department of Systems Science and Industrial Engineering (SSIE) in the Thomas J. Watson School of Engineering is expanding further and seeks five (5) Assistant Professors, one (1) Endowed Full Professor and one (1) Research Scientist (within the Watson Institute for Systems Excellence, an institute of advanced studies). Three of these positions seek research experience that complements the Binghamton University Transdisciplinary Areas of Excellence* (TAEs: The following positions are sought to start in Fall 2018: • Assistant Professor – Healthcare Systems Engineering – 3 positions* • Assistant Professor – Advanced Manufacturing – 2 positions • Endowed Full Professor – Systems Science – 1 position Positions involve: teaching at all levels (undergraduate courses through advanced graduate courses); establishing externally funded theoretical and applied research in the respective disciplines. The ideal candidates must have (i) an earned doctorate in industrial engineering or the applied sciences, (ii) excellent leadership skills, and (iii) superb records of research, including garnering funding and scholarly publications. Teaching experience is preferred. SSIE offers a BS degree in industrial and systems engineering (ISE) and MS and PhD degrees in Systems Science (SS) and ISE. Attracting over $3M dollars in research funding per year, our established faculty of 21 professors works collaboratively with over 30 sponsors from industry and federal agencies. With about 250 undergraduate, 250 masters and 120 doctoral students, the department is rapidly growing in numbers and in its reputation. SSIE also offers a cutting edge executive health systems program in Manhattan. More details are available at For more details of each position and to submit the application online, please visit Research Scientist applicants should email for more details. Review of applications begins February 17, 2018 given recent additions and will continue until each position is filled. The State University of New York at Binghamton – Binghamton University is an Equal Opportunity/Affirmative Action Employer Women and Historically Underrepresented Minority Applicants are Strongly Encouraged to Apply 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. Specific to engineering, the school is now number 7 in the Princeton Review’s list of “Colleges that Create Futures,” ranked 3rd for universities most friendly to international students, 8th best overall value for public universities and 15th best value overall for public and private schools, 8th for universities with more cultures on campus than majors, and included in the Forbes India list of 25 best-value American universities. The university is welcoming and dynamic place that continues to excel. 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.

February 2018


ORMS Today

| 59


SPECIAL ADVERTISING SECTION | View Classifieds Online at:


Gary C. Butler Endowed Chair The H. Milton Stewart School of Industrial & Systems Engineering (ISyE) at Georgia Institute of Technology is searching for an internationally recognized researcher to occupy the Gary C. Butler Endowed Chair. Criteria for evaluating candidates include: an earned doctorate in a field relevant to industrial & systems engineering, a successful research program, a strong desire and a demonstrated ability to attract external funding, facilitate corporate relationships, and provide leadership within ISyE, Georgia Tech, and the professional community, and a commitment to contributing to undergraduate and graduate educational programs of the school. The research area for this position is open. Expectations of holding this appointment include teaching graduate and undergraduate courses, providing an annual stewardship report coordinated by the ISyE Development Office and building and maintaining a relationship with the donor(s) as appropriate. The H. Milton Stewart School of Industrial & Systems Engineering is the largest such program in the United States and has consistently been ranked number one by U.S. News & World Report. Applications will be considered until the position is filled. Applicants should send a letter of application, current CV, and a list of five references to: Ms. Karen Cannon H. Milton Stewart School of Industrial & Systems Engineering Georgia Institute of Technology 755 Ferst Drive, NW Atlanta, GA 30332-0250 Georgia Tech is a unit of the University System of Georgia and is an Equal Education/Employment Opportunity Institution; applications from women and underrepresented minorities are strongly encouraged.

UNIVERSITY OF PITTSBURGH S WA N S O N S C H O O L O F E N G I N E E R I N G The Department of Industrial Engineering at the University of Pittsburgh invites applications for one or more tenure-track faculty positions. Candidates at all levels will be considered, subject to appropriate qualifications. Applicants must have strong methodological training in one or more traditional areas of industrial engineering and be motivated by impactful engineering problems in areas such as operations, supply chains, healthcare, energy and manufacturing. We are particularly interested in candidates who have the ability to conduct cutting-edge, interdisciplinary research in fields such as data analytics, machine learning, cybermanufacturing and cyberphysical systems. For junior candidates, our primary search criterion is research potential. Senior candidates must have established an outstanding research record commensurate with rank. All candidates should have evidence of, or potential for, teaching excellence. The Department of Industrial Engineering is currently comprised of 19 full-time faculty members and enjoys an outstanding reputation in a wide variety of research areas. The department maintains vibrant programs at the undergraduate, masters and doctoral levels, offers excellent laboratory facilities, and benefits from many contacts with regional and national corporations. Additional information about the department can be found at Applicants should e-mail a curriculum vitae, representative publications, and a list of at least three professional references to Review of applications will begin immediately and will continue until the position is filled. The University of Pittsburgh is an affirmative action, equal opportunity employer. Candidates from underrepresented groups are particularly encouraged to apply.

Faculty Positions Department of Industrial Engineering and Decision Analytics

The Department of Industrial Engineering and Decision Analytics invites applications for substantiation-track faculty positions in the area of (i) Predictive and Prescriptive Analytics, (ii) Healthcare and Sports Analytics, (iii) Financial Engineering and Fintech, and (iv) Demand and Supply Analytics. Applicants must have a PhD degree in Industrial Engineering, Operations Research, or a closely related area. The appointee is expected to demonstrate strong potential for effective teaching and promising research in the respective fields. Appointments at all ranks (Assistant Professor/ Associate Professor / Professor) will be considered. Salary is highly competitive and will be commensurate with qualifications and experience. Fringe benefits include annual leave, medical and dental benefits. Housing benefits will be provided where applicable. Appointment at Professor rank will be on substantive basis. Initial appointment for Assistant Professor/Associate Professor will normally be made on a 3-year contract. A gratuity will be payable upon successful completion of contract. Strong candidates applying for Associate Professor position may also be considered for appointment on substantive terms. Applications with a full C.V.; statement of research and teaching; transcript of graduate work; copies of 2 research publications; names, emails and addresses of at least three referees, should be directed to the Search and Appointment Committee, Department of Industrial Engineering and Decision Analytics, The Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong [email : Review of applications will start immediately and continue until the positions are filled. More information about the Department can be found at

H T T P : / / W W W. A N A LY T I C S - M A G A Z I N E . O R G



Anxiety over AI Time to alleviate fears of artificial intelligence

ALSO INSIDE: • • • •

Edge IoT analytics Analytics at Kroger Self-service analytics Get to know your data

Analyze This! Assignment experiment pushes Vijay Mehrotra out of his comfort zone

Check out the January/February 2018 Issue of Analytics Now Available at:

60 | ORMS Today


February 2018


SPECIAL ADVERTISING SECTION | View Classifieds Online at:

Disney At Disney, we‘re storytellers. We make the impossible, possible. We do this through constantly collecting, organizing and analyzing all forms of data allowing our analysts to bring stories to life and narrate the future of Disney experiences. We are a global, diverse team that drives business results and strategies using advanced analytics. In a rapidly changing consumer and market environment, with ever increasing touch points and data, our emphasis is on leveraging innovative and optimal solutions to provide actionable insights.

Meet Shelley | Senior Manager, Business Intelligence My team enables the business from a technology standpoint, modernizing the technology they have available to gain insights from their data. Be it a data platform that makes the data easily available for the insights, or the end-user capabilities from a business intelligence tool allowing them to access the data and create visualizations. We enable other teams within Disney to create the analysis they need to support their business. I think one of the best things about Disney is that it is a very diverse company; it’s really like working in a small city. We have everything from utilities and animals, to movies and consumer products, and everything that it takes to run all of that. All of this diversity of work allows for your career path to go anywhere. You can choose to stay in a certain path, or you have opportunity to do something completely different. I don't think you can find a better place to go. You can go anywhere to work in Business Intelligence, but when you get to create reporting on franchises, movies, TV shows and theme parks, it adds that extra element of fun!


Disney Careers

Disney Careers


The Walt Disney Company

February 2018


ORMS Today

| 61

INDUSTRY News Frontline Systems releases XLMiner SDK V2018 for highperformance predictive analytics Frontline Systems has released XLMiner SDK V2018, a next-generation version of its Software Development Kit for data mining, text mining, forecasting and predictive analytics. XLMiner SDK offers application developers working in C++, C#, Java, Python or R a powerful, high-level API for quickly creating applications that use predictive analytics. “Data mining and machine learning software has proliferated, but there’s a difference between common libraries and truly robust, high-performance software – especially if you’re working in C++, C# or Java,” says Daniel Fylstra, Frontline’s president and CEO. “XLMiner SDK is a toolkit that developers can count on to build commercial-grade applications.” XLMiner SDK provides full API support for five popular programming languages: C++ 11 or later, C# 4.0 or later, Java 8, Python 2.7 or 3.6 (both are supported), and R 3.4. In Microsoft Visual Studio and R Studio, developers will benefit from automatic recognition and “command completion” for XLMiner’s objects, properties and methods. And the new SDK is ready for REPL (Read-EvalPrint-Loop) style execution with C# Interactive. XLMiner SDK’s R support uses R-native types, including R’s own DataFrame type; hence it can be used easily with a wide range of R packages from CRAN. XLMiner SDK provides its own “R package” that can be loaded with one command from R itself or an IDE such as R Studio. For C++, C# and Java developers, XLMiner SDK should be especially welcome, since quality data mining tools have been hard to find for these popular languages. But even R and Python developers will find that XLMiner SDK offers a far better integrated, comprehensive data mining and text mining toolkit. XLMiner SDK has built-in tools to read data from SQL databases using ODBC (Open Data Base Connectivity), with special support for Oracle, Microsoft SQL Server and Access databases; OData (“the web successor to ODBC”) data 62 | ORMS Today


February 2018

sources exposing a REST API; JSON (JavaScript Object Notation); and CSV (Comma Separated Value) and Excel files. The SDK also handles unstructured text data and provides stemming, term normalization, vocabulary reduction, creation of a term-document matrix, and concept extraction with latent semantic indexing. It even has built-in facilities to draw a statistically representative sample from an Apache Spark Big Data cluster, running a Frontline-supplied component on one of the cluster nodes. Thompson Street Capital Partners makes significant investment in Gurobi Optimization Gurobi Optimization, LLC (Gurobi), provider of a leading math programming solver, announced that Thompson Street Capital Partners (TSCP), a private equity fir m based in St. Louis, has made a significant financial investment in Gurobi to help it to continue its growth as a leader in the rapidly expanding prescriptive analytics space.Terms of the deal were not announced. Gurobi’s commitment to industry-leading performance, ease of use and customer support have made the Gurobi Optimizer today’s solver of choice for more than 1,500 companies in more than 40 industries. Used in mission-critical applications or as a distributed optimization service, Gurobi allows users to state their toughest business problems as mathematical models, and then automatically considers billions – or even trillions – of possible solutions to find the best one. Thompson Street’s investment allows Gurobi to further accelerate growth while maintaining its focus on customer success. “We’re very excited about the opportunities that come with working with Thompson Street,” says Ed Rothberg, CEO and co-founder of Gurobi. “This investment gives us access to both the capital and business expertise to stay focused on our users’ needs today while developing the capabilities they will want in the future.” Craig Albrecht, managing director, TSCP says, “We like to partner with companies that are already industry leaders with proven go-to-market strategies. We are thrilled to partner with Gurobi’s superb

leadership team to help them accelerate the company’s already outstanding growth without losing sight of what has made the company so successful.” AIMMS adds ODH-CPLEX to its Prescriptive Analytics Platform AIMMS announce that it has formed an agreement with Optimization Direct to include their new product, ODHCPLEX, in the AIMMS Prescriptive Analytics Platform as an available add-on. The AIMMS Prescriptive Analytics Platform is the technology of choice for operations research and analytics professionals to build and deliver solutions that improve business performance. The Platform leverages the power of mathematical optimization and modeling to provide companies with a competitive edge and quantifiable results. AIMMS has been doing this for nearly three decades, and always tries to find innovative ways to improve value for customers. Large-scale optimization enables decision-makers to evaluate scenarios in any par t of the business. For example, supply chain organizations can use optimization for production planning and scheduling, supply chain management, sourcing, etc. As complexity and data volumes increase, the need for more powerful optimizers (so-called mathematical solvers) also increases. Optimization Direct is focusing on improving the performance of optimizers, specifically CPLEX, by offering additional mathematical “horsepower” to solve the largest and most complex planning, scheduling and supply chain optimization problems. ODH-CPLEX is a new solver designed to run on modern multiprocessor machines. “We decided to add ODH-CPLEX in our suite because the product meets the needs in specific situations where our customers require an even higher perfor mance from CPLEX. It is a great and easy way to generate value quickly,” says Gertjan de Lange, SVP Connecting Business & Optimization at AIMMS. “We have worked very closely with Optimization Direct to get the best performance from ODH-CPLEX. Their support is superb, and they allow us to set a new bar in the world of prescriptive analytics technology.” ORMS



Doug Samuelson

The gymnasts’ parable The office conversation about the Super Bowl, played the previous day, was finally winding down. As usual, there were a few complaints about allegedly incorrect officiating and odd play calls. This year, though, there was also some discussion of the league’s ­and other authorities’ ­apparent tendency to understate the risks of injury to players. “There’s a foundation, led by former players, lobbying to get tackle football banned for kids under 13,” Joe pointed out, “but even school districts don’t seem to want to do it. How many more of these CTE cases do they need? The NFL and the big-time college conferences check players for concussions, but nobody checks players, especially young kids, for possible damage from repeated less violent hits. That’s true even though now there’s research indicating that the number of hits is more important than the severity. And it has a worse effect on younger people.” “At least the NFL is doing something,” Sheila countered. “Where are the policy changes from the NHL, the boxing federations, Mixed Martial Arts and NASCAR? And how about the way USA Gymnastics protected that doctor who was sexually abusing young girl gymnasts? Doesn’t any organization take responsibility for endangering its people?” Barry smiled a bit sourly. “Not all that often, I’d say. But I suspect the former gymnasts will end up owning USA Gymnastics by the time the court cases are done,” he said. “I don’t imagine that association has anywhere near enough liability insurance.Then we’ll see whether the women protect their younger colleagues. I’m not sure which way I’d bet.” Sheila was indignant. “Why wouldn’t they?” she demanded. Barry replied, “Remember ‘Animal Farm’? Lots of organizations start out with equality and cooperation, but they evolve into self-protection for the top 64 | ORMS Today


February 2018

managers. And when you think about how it happens, it’s logical, and hard to prevent. People get invested in the reputation of the organization and want to protect it from any kind of harm, including legal problems and bad publicity. And besides, they believe that what the organization does is mostly good, so they have a hard time accepting that it did or condoned something bad.” “But why didn’t the parents do more to force the issue?” Sheila insisted. “Don’t you think families would be more involved from now on?” “Maybe, maybe not,” Barry shrugged. “You have to remember that these athletes, like a lot of young top-level athletes, get to the top by immersing themselves in a ‘hothouse’ training environment, away from their families, away from most friends, away from most activities. They get used to doing whatever their coaches and trainers tell them. “And,” Barry added, “you don’t get to succeed at that level of sports, any sports, without a really high pain tolerance. Remember Kerri Strug, the girl who landed the vault that won the team gold medal for the U.S. in 1996? How she landed it on one foot, after a bad injury on her first vault?” The others nodded. Barry said, “She was on ‘The Tonight Show’ shortly after the Olympics. Jay Leno asked her, ‘I heard or read somewhere that you heard this loud crack from your ankle when you landed that first vault. Didn’t that tell you something serious was wrong?’ And Kerri Strug, calm as could be, replied, ‘Gymnasts hear pops and cracks all the time.’ Now she’s out on the circuit as a motivational speaker, encouraging people to work through adversity to pursue their passions. I’m guessing the parents mostly bought into this approach, too. “My daughter took years of ballet,” Barry continued. “I remember one of her

doctors telling her, ‘Look, let’s be realistic. Dance careers usually end in injury. Major dance careers end in major injury. If you just have to dance, you keep doing it working through pain until you simply can’t any more. Look at the great retired dancers who appear at awards ceremonies and such events. Most of them need canes to walk. Get the picture?’ And that, by the way, is why eventually my daughter and I had a sad conversation in which she decided, with my strong encouragement, to give up competitive dancing.” “Sounds like a good choice,” Sheila acknowledged. “You have to balance attaining your goals against new information about what that might cost.” “Yeah,” Barry smiled, a bit more grimly. “Now let’s try telling that to our political leaders. With all due respects to Vince Lombardi, who’s famous quote, ‘Winning is the only thing,’ might be a good value to build a football team around, but in a professional association or community group, excessive competition and defensive management can be fatal, with lots of bad consequences for people in the organization. And it’s a really harmful core value when you’re trying to run a whole society and leaders decide ‘my group simply has to win.’ How many times do we have to see it before we recognize the pattern?” “But what can we do about it?” Joe asked. “I’m not entirely sure,” Barry replied somberly, “but I’d start with what the experts on domestic abuse I know have told me: The first step to opposing abuse is to be quite sure that it really is abuse, and you really don’t deserve it. Get out of there, any way you can! If you can’t get out, enlist allies and organize resistance. Don’t let yourself get talked into believing you somehow asked for it. And when you’re in a situation where you have the power, don’t do unto others what you didn’t like having done to you! And if people complain about how they’re being treated, pay attention!” ORMS Doug Samuelson (samuelsondoug@yahoo. com) is president and chief scientist of InfoLogix, Inc., a small R&D and consulting company in Annandale, Va.










Thank you for supporting your profession in 2018 with your INFORMS membership.




CONTINUED Online renewals available at 443.757.3500 or 1.800.446.3676











THE GAMS COMMUNITY Find your way to stay in touch at Connect with other GAMS users and exchange experiences Get help and tips!




Newsletters & Mailinglist


Solution Specialists Blog

Stay up-to-date on important GAMS news

Find new job opportunities, new employees, or solution specialists

Become a part of our large, global GAMS community!

High-level algebraic modeling system which incorporates all major commercial and academic state-of-the-art solvers Phone: (202) 342 0180

Orms Today - February 2018  
Orms Today - February 2018