Previous month:
September 2015
Next month:
November 2015

October 2015

#INFORMS15 Undergraduate O.R. Prize Sessions

Both sessions on Sunday (11am and 1:30pm) will be held at the Marriott in Room 403. They provide a fantastic opportunity to discover the work today's best undergraduates do in operations research - those students will certainly be among tomorrow's best graduate students or best young practitioners of O.R. 

The talks in each session with their abstract are provided in the attached files (Session 1 and Session 2) since copying and pasting the online schedule runs into formatting issues and makes the abstracts unreadable.

I am sure it would mean a lot to the undergraduate students if you could take the time to drop by. 

PLEASE note that Paige von Achen's talk has been moved to 1:30pm, which is correctly reflected in the online schedule but not the print schedule.

Below, the finalists are listed in alphabetical order of the first author. 

  • Beril Burcak, Osman Raug Karaaslan, Ayca Karatape, Alaz Ata Senol, Hakan Senturk and Kaan Yavuz, "Integrated optimization of aircraft utilization and on-time performance", advised by Prof Kemal Goler at Bilkent University 
  • Massey Cashore, "Multi-Step Bayesian optimization for one-dimensional feasibility determination", advised by Prof Peter Frazier at Cornell
  • Kyle Cunningham, "Alleviating competitive imbalances in NFL schedules: an integer-programming approach", advised by Prof Murat Kur at SUNY Buffalo
  • Omar El Housni, "Piecewise static policies for two-stage adjustable robust linear optimization problems under uncertainty", advised by Prof Vineet Goyal at Columbia University
  • Pengyu Qian, "A composite risk measure framework for decision making under uncertainty", advised by Prof Zizhuo Wang at the University of Minnesota - Twin Cities and Prof Zaiwen Wen at Peking University.
  • Chao Qin, "A faster algorithm for the resource allocation problem with convex cost functions", advised by Prof Cong Shi at the University of Michigan
  • Magdalena Romero, "Optimal resource allocation in breast cancer screening with different risk groups", advised by Prof Qingxia Kong at the Universidad Adolfo Ibanez in Chile
  • Cagan Urkup, Ezgi Karakas, Fehmi Mert Gurel and Kaan Telciler, "Routing optimization of a drone for agricultural inspections", advised by Prof Sibel Salman at Koc University
  • Paige von Achen, "Optimizing community healthcare coverage in remote Liberia", advised by Prof Karen Smilowitz at Northwestern University
  • Andy Zheng, "Robust multi-objective clustering", advised by Prof Omid Nohadani at Northwestern University. 

#INFORMS15 TutORial 1: Data-Driven Stochastic Programming using Phi-Divergences

We have an exciting line-up of tutorials for INFORMS 2015! As attendees are beginning to arrive for the conference, Dionne Aleman and I (2015 TutORial co-chairs) would like to draw your attention to the excellent tutorials we have scheduled. We are grateful to our speakers for making the time to prepare engaging presentations on cutting-edge topics.

The first talk will be on Data-Driven Stochastic Programming using Phi-Divergences by Guzin Bayraksan, tomorrow at 8am in Room 108A. 

Abstract

Because collected data is variable, and oftentimes it is used to forecast model parameters (where forecasts themselves contain uncertainties), at the heart of data-driven optimization is optimization under uncertainty. Stochastic programming provides a way to write mathematical optimization problems by explicitly considering the uncertainty in the problem parameters by incorporating random variables and probabilistic statements.

Most of classical stochastic programming assumes that the distribution of the uncertain parameters are known, and this distribution is an input to the model.  In many applications, however, the true distribution is unknown.  If the distribution used in incorrect, a stochastic program can give incorrect solutions. An ambiguity set of distributions can be used in these cases to hedge against the distributional uncertainty.  Such a set of distributions can be formed using various sources of data.

Phi-divergences (e.g., Kullback-Leibler divergence, chi-squared distance, etc.) provide a measure of distance between two probability distributions.  They can be used in data-driven stochastic optimization to create an ambiguity set of distributions that are centered around a nominal distribution.  The nominal distribution can be determined by collected observations, expert opinions, results of expensive simulations, etc.  Many phi-divergences are widely used in statistics; therefore they provide a natural way to create an ambiguity set of distributions from available data and expert opinions.

In this tutorial, we present two-stage stochastic models with distributional uncertainty using phi-divergences and tie them to risk-averse optimization. We examine the value of collecting additional data. We present a classification of phi-divergences to elucidate their use for models with different sources of data and decision makers with different risk preferences. We illustrate these ideas on several examples. 

Bio

Güzin Bayraksan is an associate professor in the Integrated Systems Engineering Department at the Ohio State University. Prior to joining OSU, she was a member of the Systems and Industrial Engineering Department and the Graduate Interdisciplinary Program in Applied Mathematics at the University of Arizona. She received her Ph.D. in Operations Research and Industrial Engineering from the University of Texas at Austin and B.S. in Industrial Engineering from Bosphorus (Bogazici) University in Istanbul, Turkey.  Her research interests are in stochastic optimization, particularly Monte Carlo sampling-based and data-driven methods for stochastic programming with applications in water resources management. She is the recipient of 2012 NSF CAREER award, 2012 Five Star Faculty Award (UA), and the 2008 INFORMS best case study award. She currently serves as the president of the Forum for Women in Operations Research and Management Science (WORMS), an elected member and treasurer of the Committee on Stochastic Programming (COSP), and on the editorial board of IIE Transactions.


INFORMS 2015 TutORials volume

IMG_2921I found the printed version of the INFORMS 2015 TutORials book in my mailbox today! (See picture on the left.) I co-edited it with Dionne Aleman of the University of Toronto. It is entitled "The Operations Research Revolution". We have some amazing tutorials lined up by experts in our field such as Laura McLay, Guzin Bayraksan, David Goldsman, Erick Delage, Dan Iancu, Art Chaowalitwongse, Josh Taylor, Meinolf Sellmann and more, and I can't wait to hear their presentations at the annual meeting. The papers will be available for free to INFORMS members. The link isn't up yet but I'll make sure to post an update as soon as I have more information. Dionne and I are looking forward to an informative set of presentations on cutting-edge O.R. topics in just a few weeks in Philadelphia!


Rotman Management Magazine

12315I finally got around to buying (and reading) a copy of Rotman Management Magazine,the magazine of the Rotman School of Management at the University of Toronto, which I noticed in newsstands some time ago next to Harvard Business Review, MIT Sloan Management Review, Strategy+Business and Stanford Social Innovation Review but never picked up. While SSIR has found a niche for itself (social innovation) that makes it complement market leader HBR without competing with it, and the MIT name is strong enough to find its own set of devoted readers, Strategy+Business could use a redesign of the magazine's inner pages and I was curious to see how RMM would position itself in a very competitive field. 

If one has to give credit to a single person for the emergence of Rotman as a powerhouse business school with its own management publication, global reputation and a ranking as one of the world's best business schools (#1 business school in Canada, #4 in faculty research worldwide, #4 PhD Program, Top 10 in finance), that person would be Roger Martin, who was Rotman's Dean from 1998 to 2013, has popularized concepts such as integrative thinking and design thinking and is the author/co-author of best-selling books such as The Design of Business (2009) and Playing to Win (2013). As Dean, he also oversaw the launch of RMM back in 1999.

While part of Rotman's rising worldwide reputation is certainly due to Toronto's success story as a magnet for the creative class - to use a term first introduced by Richard Florida (now head of the Martin Prosperity Institute at Rotman) - there is no doubt that the growing impact of Rotman's faculty and graduates can also be attributed to an outstanding marketing strategy, in which RMM plays a critical role.

In contrast with HBR, which comes across as rather independent of Harvard Business School, in the sense that it does not try to specifically push research of HBS professors, RMM makes no mystery that it is Rotman's magazine. (As it should, if the overarching goal is to position Rotman as one of the very top schools of management in the world.) If you miss the words on the cover, you will quickly realize it due to the large proportion of contributors or interviewees affiliated with Rotman, and the way a number of them drop the Rotman name in their articles.

For instance, Jim Fisher in The Thoughtful Leader: A Model of Integrative Leadership mentions he teaches a course on leadership at Rotman in the first column of the article's first page. (The article discusses Managing, Directing and Engaging models of leadership. It should be quite obvious from their names which one the author advocates for the future, but the article was thoughtful and well-presented.) 

Dilip Soman, in The Last Mile: Using Behavioral Insights to Create Value, advocates that "leaders spend too much time on 'first-mile' issues like strategies and too little time on the 'last mile', where consumer decisions are actually made". He is a professor of marketing at Rotman. 

Bernardo Blum, Avi Goldfarb and Mara Lederman provide a Path to Prescription: Closing the Gap Between the Promise and the Reality of Big Data. If the "Path to Prescription" part of the title made you expect something about big pharma, you might want to know the "prescription" refers to prescriptive analytics, as opposed to its descriptive or predictive counterparts. All three authors are affiliated with Rotman.  

Eileen Fischer and Rebecca Reuber author Retweet This: The Power of a Multi-Dimensional Approach to Social Media. The second author is a Professor of Strategic Management at Rotman. 

Jonathan Bailey and Tim Koller investigate the effectiveness of boards of directors in All Aboard: Making Board Effectiveness a Reality. They ask: "Is your board of directors effective? A veteran director finds many boards wanting - and considers how to improve them." That veteran director is David Beatty of the Rotman School of Management, whom they interview at length. (Bailey and Koller are with McKinsey.)

Rotman's new Dean Tiff Macklem is the subject of a thoughtful interview on his first ten months on the job. There is also a great interview with Rotman alum Richard Nesbitt MBA'85 now adjunct faculty at Rotman who provides leadership lessons from a financial services veteran. Partha Mohanram, the CPA Ontario Professor of Financial Accounting, opens up the black box of accounting. Mine Moldoveanu, founder and director of the Mind-Brain-Behaviour Hive at the University of Toronto and vice-dean for learning and innovation at Rotman, discusses how to harness brain science and wearables for personalized learning. 

But it is important to note that the contributions of Rotman-affiliated professors are interspersed with excellent interviews of management thought leaders with no connection to Rotman or Canada (that I'm aware of): Thinkers50 member Linda Hill on collective genius, four leading management thinkers in a Leadership Forum on the architecture of management (Herminia Ibarra, Georg Polzer, Tammy Erickson and Vineet Nayar), authors affiliated with MIT Sloan Management Review and the Boston Consulting Group who wrote on collaborative leadership for sustainability, MIT faculty Erik Brynjolfsson and Andrew McAfee on The Digitization of Just About Everything.

My favorite articles were Portrait of a Leader: Sheryl Sandberg by Wharton's Stewart D. Friedman, who provides a useful insert on his model of total leadership, which he describes in more detail in his book of the same name and The Power of Optimism by Bill and Melinda Gates. ("If you want to do the most, you have to see the worst." It was a beautiful article, although I wonder what happened to the patient dying from AIDS after Melinda Gates carried her onto the rooftop so that she could see the sunset, at the dying patient's request, because employees refused to do so. The employees' reason was that she would soon be dead and the other patients were keeping them busy. Melinda Gates makes it clear she told the employees where the patient was so that they would pick her up after the sunset was over, because Gates couldn't stay that long, but part of me wonders if they didn't just let her die on the rooftop after Gates was gone. Read the whole article and tell me what you think.)

I also enjoyed Co-creating the future: The Dawn of System Leadership by Peter Senge, Hal Hamilton and John Kania. (No Rotman affiliation there: Senge is with MIT, Hamilton with the Sustainable Food Lab and the the Academy for Systemic Change and Kania with FSG.) A longer version of this article first appeared in SSIR. How to lead at your best, by Joanna Barsh and Johanne Lavoie, was excerpted from their book on Centered Leadership. (Hints: (1) Practice the pause. (2) Forge trust. (3) Choose your questions wisely. (4) Make time to recover.) They are with McKinsey. Change guru John Kotter of HBS discusses the importance of agility and provides an overview of management vs leadership in a thoughtful Q&A.

I very much liked the Q&A with famed executive coach Fred Kiel, who defines leadership character and defines a Return on Character matrix, and the Q&A with consultant/author Mark Barden, who explains "why we need to change our opinion about constraints." He defines three types of attitudes with respect to constraints: the Victim, the Neutralizer and the Transformer. ("These are people who see the constraint as an opportunity.") A key insight, for which he thanked Pentagram designer Michael Beirut, is that those are three stages all of us go through rather than three different types of people

Canadian-born Brad Katsuyama worked for many years for the Royal Bank of Canada and is now the founder of IEX, an alternative stock market. His adventures in high-frequency trading made him a central character in Michael Lewis's book Flash Boys.  

In other words, RMM is (rather) careful about pushing the Rotman brand name while mixing Rotman articles with those by world-renowned thought leaders of peer institutions or aspirational peers. What can other schools learn from this example? Most universities, colleges and sometimes even departments now publish regular issues about research advances by their faculty. Those publications are often sent to alumni or donors ("members of the family", as Michael Kaiser would say about performing arts organizations). They usually push news updates that will only be meaningful to people who already share an affiliation with the school. RMM, though, takes the next step in providing leadership news and business insights in a useful, actionable format. It so happens that many of those new models or insights are contributed by Rotman faculty. This is a great way to position the Rotman School of Management on an international stage. I'll be looking forward to the next issue in the winter.


Physicist Brian Greene at Lehigh

BrianGreeneLehighYesterday I attended the public talk by theoretical physicist and string theorist Brian Greene at Lehigh, which was held as part of the celebrations for the university's 150th anniversary. The auditorium at Zoellner Arts Center was packed with students, professors and members of the general community. The central theme of Greene's talk (as I understood it) was that at any point in time, innovations and discoveries are made by people who have been able to keep a child-like ability to look at the world anew, and their predecessors, who made their own innovations and discoveries, might be their fiercest opponents in not recognizing the validity of their younger colleagues' claims. Greene motivated his talk through the example of Albert Einstein, who pioneered general relativity (Greene pointed out that November 25, 2015 will mark the 100th anniversary of general relativity, by the way) but then didn't believe Karl Schwarzschild when the latter first wrote that Einstein's theory led to black holes. In a way, Einstein had lost the ability to see how far his own theory went. 

Greene then translated this issue to present times when he argued in favor of the holographic principle, which suggests that we are 3D projections of a 2D universe. (You can read more about his theory in this 2012 Wired article.)  This principle arose from the resilience of information in black hole that apparently follows from Einstein's principle. When an object enters a black hole, its information is smeared over the boundary of the black hole and thus the object can be reconstituted.

Greene had three main stories related to Einstein, prefaced by one about the sense of wonder of a second grader. Somewhere along the way he touched upon the issue of universe vs multiverse, and argued (I think, if I understood him correctly) that in many years, if the universe keeps expanding, we would have very dark skies that would look empty but wouldn't be, and we could try to warn our future descendants but if given the choice between listening to ancestors of an era long past or believing their state-of-the-art observations, they would believe their observations, and be wrong.

He talked for one hour without notes, was an exceptionally engaging speaker who told some jokes and never stayed behind the lectern for long, except when he showed many short, excellent videos to explain the physics he was talking about. During the Q&A, he answered every single question with ease, including the more technical ones. He also made it clear that, although the holographic principle does seem outlandish (my words, not his), he has no interest in wasting his time on theories that can't possibly be true.

I enjoyed reading Greene's books The Elegant Universe and The Fabric of the Cosmos some years ago, and it turns out Greene is as good a speaker as he is a writer. Unsurprisingly, he gave a TED Talk back in 2012, which you can watch below (from the TED channel). 


Support the Pennsylvania Shakespeare Festival!

PSFThe Pennsylvania Shakespeare Festival has for mission “to enrich, inspire, engage, and entertain the widest possible audience through first-rate professional productions of classical and contemporary plays, with a core commitment to the works of Shakespeare and other master dramatists, and through an array of educational outreach and mentorship programs.” (quoted from PSF website)

Located in Center Valley, PA, about one hour north of Philadelphia, on the grounds of DeSales University, it is in a unique position to contribute to the Lehigh Valley’s revitalization and repositioning as an area attractive to many companies (due to its proximity to major markets, reputable healthcare facilities and well-known universities) by providing high-quality theater productions that will appeal to many. Another critical asset of PSF is its close relation with DeSales University, which has a superb theater program whose alumni include wonderful artists such as Marnie Schulenburg, Zack Robidas and Phoenix Best: some DeSales theater students serve as interns to the festival and others move to the Young Company (the branch of PSF that presents Shakespeare in schools) when they graduate. It has become more and more fashionable in recent years to advocate for layered mentoring, where students in any discipline learn by working side by side with more advanced peers, themselves mentored by more experienced professionals, but through its unique location PSF really gets to be at the forefront of this phenomenon.

Many PSF actors enjoy returning to the festival as many times as casting allows. (This is also a sure sign PSF is doing something right!) Philadelphian Anthony Lawton, for instance, is a regular at the festival who starred in Henry V and The Foreigner this season and Macbeth last year, among many other festival appearances. Phoenix Best, who made an incredible stage debut in Chicago some years ago in a theater department production, has been seen in Oklahoma!, Music Man, Henry V and Two Gentlemen of Verona, among others. I very much enjoy seeing some of the same actors year after year in different roles. It gives patrons the opportunity of discovering actors in different roles that showcase the incredible breadth of their skills, and creates a real feeling of "family".

In addition to producing plays and musicals of the highest caliber for the festival, PSF does tremendous work in the communities through the Linny Fowler WillPower Educational Tour, which serves schools in Pennsylvania, New Jersey and Delaware through October and early November. The play this year is Julius Caesar. The tour is the centerpiece of PSF's educational programming, recently selected to be part of Shakespeare in American Communities, sponsored by the National Endowment for the Arts in collaboration with Arts Midwest. Over 145,000 students have been exposed to Shakespeare's tremendous art thanks to PSF's Linny Fowler WillPower Educational Tour. The tour consists of a 80-minute professionally produced play with sets and costumes, a half-day or full-day of programming, including workshops, delivered by experienced teaching artists, a classroom study guide for the play and a post-performance discussion with the artists.

PSF is now seeking to prepare for the next phase of its growth and national impact by (i) replacing or upgrading equipment such as lighting equipment, wireless microphones, sound and light control boards, (ii) expanding arts programming in schools to reach more students in more depth (with a goal of bringing Shakespeare to between 30 and 50 schools each season and supporting activities such as the student Shakespeare Competition) and (iii) hiring the best artists in the industry, from actors to directors to designers, to expose PSF patrons and local students to the very best, most innovative theater available today, turning each spectacle at PSF into memories cherished for a lifetime with families and friends.

To achieve its goals, PSF launched a fundraising campaign a few months ago. It is very close to achieving its fundraising goals but fewer than one hundred days remain until the campaign ends! Please help PSF create world-class theater at the summer festival and in local schools by making a donation here.

If you have never been to PSF, the videos below provide highlights of past seasons.