``
Phoenix, AZ October 14-17


Conference Blog


by Allen Butler

What does one have to do with another?  Not much , but those are the two topics that are on my mind this morning.  Yesterday, in the Roundtable meeting, I learned all about Big Data and I even know how to pronounce Hadoop now.  The numbers these people use are ones whose names I didn’t know – a zettabyte of data???  The applications of Big Data run the gamut from high frequency financial trading to bidding for the opportunity to place those little ads on the web page you just visited.  BTW, in this field you may only have 30 milliseconds to decide whether and how much to bid, of course you get that opportunity billions of times a day.  I found all the talks fascinating.

Today I have the pleasure of chairing the WAGNER PRIZE sessions in Auditorium 3.  I may be biased, but other than the plenaries and keynotes, I think these are some of the best talks you will find at the conference.  These are the finalists, so the quality of the talks tends to be extremely high.  Even if the talk is in a field with which you are not all that familiar, I guarantee you will learn and enjoy.  And, every talk is about a real world application!  Three sessions 8-9:30, 11-12:30, and 1:30-3.  Come one, come all!

 

by Paul Rubin

…with apologies to Del Shannon.

It’s always warring emotions for me on the last day of an INFORMS conference: sorry to see the fun end and online friends become virtual again (until next year); relieved to make it all the way without my brain exploding; frightened that I’ll find out that this time I finally drank enough coffee to cause organ damage.

Thanks to George Runger of ASU for an entertaining and informative tutorial on data-mining, which drew a large audience (on getaway day, at 8:00 a.m. no less). The audience skewed young, which is a good thing given the predicted demand for “deep” data miners.

Big thanks to the conference organizers for a great meeting, which ran very smoothly from my grunt-level perspective.

Lastly, a shoutout to the INFORMS marketing and social media folks for all the effort they put into keeping us up to speed on goings on.

I’m outta here now. See you all at the next one!

by Walt DeGrange

Over the past few years the Military Applications Society has rewarded attendees and presenters at the last MAS sponsored conference session MAS coins and we take a group picture. Here is this year’s photo.

 

 

 

 

 

 

For whatever reason (hard charger, can’t get enough of the INFORMS Annual Conference, screwed up travel planning, etc.) these folks stayed to the bitter end and enjoyed the last MAS briefs of the conference. As the session chair and one of the speakers I thank all that attended and wish everyone safe travels home.

by Laura McLay

The WORMS (Women in OR/MS) Award was awarded to Vicki Sauter.

The Award for the Advancement of Women in OR/MS celebrates and recognizes a person who has contributed significantly to the advancement and recognition of women in the field of Operations Research and the Management Sciences.

Vicki is a professor in the College of Business Administration at the University of Missouri at St. Louis. Vicki is a trailblazer that achieved many “firsts:” she was the first woman in her department, the first woman to get tenure, and was the first woman to be General Chair of the ORSA/TIMS Annual Meeting in 1987 (ORSA/TIMS eventually became INFORMS). As chair, she ensured that child care was available for those who needed it. That was the only time that child care was available.

<aside>Ahem, INFORMS. It’s been 25 years, and I see lots of young operations researchers with young children at the conference!</aside>

Vicki maintains a blog about technology and is on twitter.

by Thiago Serra

ORMS non-stop from 8 AM to 10 PM (and even more with other students staying at my hostel)! Those were great days, where I learned a lot and met many amazing people. I am already looking forward to attending EURO-INFORMS 2013 in Rome. Hopefully, see you there!

by Tallys Yunes

Tuesday’s Treasures: If you missed the Constraint Programming sessions on Tuesday (TB12 and TC12), you should go back and try to get those research papers (by contacting the authors). It was beautiful: a strong display of difficult, important problems tackled effectively by powerful solution algorithms (cloud network design, scheduling, disaster recovery, offshore drilling operations). A joy to watch.

Wednesday’s Winner: The Data Mining tutorial (WA16) was very well delivered and pretty informative. I purchased the Tutorials book, but I’m glad I attended this one. There was a pretty nice collection of pointers to useful information on the topic. I also got to chat with Tim Hopper one more time and he agreed to be my go-to guru as I attempt to learn Data Mining. Thanks, Tim!

Final Farewells: As it all comes to an end, and I look back at how I spent my time in Phoenix, the only conclusion is that it was a great conference. I managed to do a bit of everything: I listened, I learned, I networked, I established new contacts, I had research meetings, I had philosophical meetings, I caught up with old friends, and I had fun. As one of my ribbons states, I was indeed delighted.

As for my video promises: I did attend my own talk, I did attend the social networking panel, I did engage in ribbon war (with help from Mary L. and Bo Jensen), someone actually asked for my papers (research ones) after my talk, I did (with pleasure) buy some people drinks, and I did meet INFORMS staff members for the first time; specifically, it was very nice to finally meet Ellen Tralongo and Shirley Mohr.

May you all have safe and uneventful travels to wherever your final destination may be. In the worst case, I’ll see y’all again in about a year!

by Paul Rubin

With four technical sessions, the Analytics Certification reception, the general reception and some stimulating discussions with colleagues, Tuesday was quite the day.

It started for me with the TA17 session on spreadsheet modeling, in which I had a chance to serve on a panel with three other presenters, including Dan Fylstra of Frontline Systems (and, more relevantly, one of the pioneers behind Visicalc, the seminal spreadsheet). My job, as cleanup hitter, was to be the resident skeptic. Dan applied a little presenter jujitsu by ending his segment with a quote from the abstract of my presentation, which he then modified to express his take (with which, oddly enough, I agree).

TB22 had the title ” Branching in Mixed-Integer Programming I”. Fascinating stuff, right? The room was pretty close to packed! When large numbers of people are geeking out on branching strategies, you must be at INFORMS.

A productive “working” lunch and two more sessions later there was highly informative information session (followed by wine and cheese, hence my presence) on the new analytics certification program. If you’re interested, keep an eye out for further communications from INFORMS via the usual channels (OR/MS Today and Analytics magazines, e-mail from the mother ship, …). I don’t think I would qualify to sit for the exam (lack of relevant experience), but that’s fine; I have the only certifications I need to be a retiree (gray hair and knees that make crackling sounds when I stand up).

That brought me to the general reception, the theme of which seemed to be “compete with Irv Lustig for biggest hat”. The food was good (and plentiful), and the band was playing my kind of music. One highlight was a discussion with an animated bartender well under half my age about the fate of the Temptations. I always enjoy meeting young people discerning enough to groove to real music.

Some stimulating conversation during and after the reception (yes, it was possible to be heard over the band … if you sat outside the ballroom} capped off the evening. I hope everyone else had as good a day – and I hope those of you sitting near the band retain some hearing.

by Guillaume Roels

I have to say that I am quite impressed with the management of the flows of people during the conference. The conference seems to be much bigger than in the previous years. There are 73 sessions per time slot, 4 time slots/ day, and 4 days of conference, which gives us a total of 1168 sessions. With 4 talks scheduled per session, that should give us more than 4,000 talks overall. This is a record, isn’t it?

If the number of attendees is proportional to the number of talks, this should then be one of the largest INFORMS conferences ever. What is impressive is that everybody I have met seems to have been able to have found a hotel room near the convention center, the lines are never too long (even at the highly attractive Starbucks after lunch), there are plenty of food options for lunch within reasonable walking distance, taxis are always available, etc.

Applying the Number 1 principle of queuing theory, it must be that Phoenix has a lot of slack capacity. This is a blessing for this conference. I have terrible memories of other past conference locations (I won’t name them) in which all nearby hotels were booked, the minimum wait for a cab was at least 1/2 hour, and restaurants were so packed at lunchtime that we had to walk several miles to find something to eat. Although Phoenix doesn’t have as many museums as New York or Chicago (but who would spend time in a museum while there are so many interesting talks going on?), I believe that access and convenience should be important criteria for choosing a particular conference location.

Access and convenience are important dimensions for service innovation. Some service companies have differentiated themselves on that dimension. For instance, Virgin Airlines is offering complimentary limo services (to eligible customers) to transfer from and to airports. Although limo services may seem to lie outside the scope of the airlines business, it is actually an integral part of the business of transporting people from Point A to Point B, which is I suppose the way Virgin positions itself. Similarly, the conference organizers could have limited themselves to offering an interesting program, but it is quite remarkable that they have also adopted a more global view on the “conference experience” and picked a location that had the capacity to host such a large crowd without creating excessive waits. Congratulations!

by Matthew Saltzman

In a remarkable display of synergy, the 2012 COIN-OR Cup award reinforced the theme of the 2012 INFORMS Impact Prize. The Impact Prize was awarded to the entrepreneurs who created the OR modeling language industry. The COIN-OR Cup recipients were the open-source developers (including some of the same people) who created tools linking the modeling language systems to the COIN-OR solver libraries.

Congratulations to the winners: Bjarni Kristjansson for CoinMP;
Marcel Hunting and Marcel Roelofs for AIMMSlinks; and
Michael Bussieck, Steven Dirkse, and Stefan Vigerske for GAMSlinks. In addition, Pierre Bonami received an Honorable Mention for the Bonmin MINLP package.

The COIN-OR Cup Judges were: Andrew Mason (OpenSolver); Iain Dunning (OpenSolver); Stu Mitchell (PuLP); and Kipp Martin (Optimization Services and COIN-OR Foundation member).

Congratulations to Bjarni also for his recent engagement.

The citation follows.

After careful consideration of many excellent submissions, and much difficult deliberation, the COIN-OR Cup Judges have agreed on a 2012 COIN-OR Cup winner. The 2012 INFORMS COIN-OR Cup is awarded to a joint nomination of the managers and friends of three COIN-OR interfacing projects:

Bjarni Kristjansson for CoinMP,
Marcel Hunting and Marcel Roelofs for AIMMSlinks, and
Michael Bussieck, Steven Dirkse, and Stefan Vigerske for GAMSlinks.

In their deliberations, the committee members praised this team for the combined contributions they have made over many years to facilitate increased access to the COIN-OR solvers.

Bjarni Kristjansson has been developing and supporting CoinMP for 7 years now. COIN-MP provides easy access to COIN-OR solvers, such as CLP and CBC, from a variety of programming languages, including Bjarni’s own MPL Modeling System. A significant milestone for COIN-OR and CoinMP was the recent adoption of CoinMP for use in OpenOffice.

The COIN-OR AIMMSlink project was created in 2009 to build on and improve earlier work to integrate COIN-OR solvers with the AIMMS modelling system. Marcel Hunting and Marcel Roelofs both contributed to interfacing IPOPT and CBC with AIMMS. From 2010, these solvers have been distributed with the AIMMS software, making them available to all AIMMS users.

GAMS support for COIN-OR software dates back to Michael Bussieck’s work in 2004 with CBC. This work has continued and been extended so that GAMMS now also supports OSI, Ipopt, Bonmin, Couenne, SCIP, the Branch-Cut-Heuristic, and the Optimization Services project running on Linux, Solaris, Mac OS, and Windows. Members of the judging committee who have personally used GAMSlinks and found it very useful.

GAMS, AIMMS and the MPL Modeling System are significant players in the modeling arena and having interfaces to the COIN-OR solvers gives the COIN-OR solvers tremendous exposure.

During their development work, this team has also contributed to COIN-OR with solver improvements, bug reports and bug fixes. The judges were very pleased to learn that COIN-OR solvers are tested daily during GAMS and AIMMS regression runs on multiple platforms. It is thanks to the many years of high quality contributions from teams such as this that COIN-OR is so successful today.

The judges also wish to acknowledge the nomination by Andrea Lodi of Dr. Pierre Bonami as the main player in the development and contribution of Bonmin. It didn’t take much googling to see that Bonmin is making a huge impact in the optimisation community. Pierre’s most recent contribution is a nice piece of code that improves the performance of the Outer Approximation algorithm within Bonmin in the special case of Separable (Convex) MINLPs. The judges wish to formally recognise Pierre’s work by recording his contributions as being Highly Commended.

by Ken Chelst

Our next one day MINDSET workshop is scheduled for Queens, NYC. We already have 40 high school math teachers signed up for the program.

by Ken Chelst

Karl Kemp, Senior Scientist of Intel, brought sample products for the high school teachers to see up close. He talked about several of his OR projects, one of which saved more than billion dollars and another that generated increased revenues of more than a billion dollars. One teacher walked away saying she could now tell her students she had an operations researcher who used mathematics to save a billion dollars.

by Ken Chelst

Dave Goldsman, James Cochran, Ken Chelst and Tom Edwards delivered a warmly received workshop from 8 am to 3 pm to 72 Phoenix area teachers. They learned about MCDA, Excel Solver for LP, Queueing and Simulation. They walked away committed to bringing the material into different math levels of their high schools. We expect to hear from several schools about bringing an entire semester into a 4th year math course in Fall 2013.

by Barry List

For INFORMSers like me who work inside a bubble – I’m Director of Communications for INFORMS, and I spend most of my time talking and corresponding with volunteer leadership and staff – INFORMS TV has proven an unexpectedly satisfying experiment in putting a voice and a face to those who INFORMS members know principally as a name and a distant presence.

INFORMS TV has been an experiment for the association, which accepted a bid from the company Websedge to prepare thought leadership spots for organizations that work or teach O.R./analytics, and also add interviews with INFORMS officers, volunteers, staff, and members.
For me, the thought leadership spots at schools like the Naval Postgraduate School and the University of Alabama, and companies like Lockheed Martin gave me the opportunity to take a look inside, down the hallway, into the computer lab, the classroom, and the workstation. Most important, it provided a firsthand look – a virtual look of course – at people who are pioneering programs or conducting exciting research.

For example, I exchanged emails many months ago with Dean J. Michael Hardin of the University of Alabama’s Culverhouse College of Commerce about the business intelligence/analytics program at the school, but didn’t have the chance to meet him in Huntington Beach, at a meeting of analytics school representatives. INFORMS TV let him explain the work of his campus more vividly than a website or email could do. I got to know him better.

I recorded a podcast with David Alderson of the Naval Postgraduate School about the security threats to America’s infrastructure, but watching him onscreen gave a more complete sense of his personality than I got recording the interview. It was also great to see his colleague Gerald Brown on camera for NPS (I’ll always be grateful to Jerry for making possible the Science of Better podcast that ORMS Today’s Peter Horner and I recorded with former Chairman of the Joint Chiefs of Staff Admiral Mike Mullen).
Hopefully, INFORMS members came away from INFORMS TV with a more tangible sense of INFORMS leaders than they had before. You may have read an interview with INFORMS Executive Director Melissa Moore or a column by INFORMS President Terry Harrison, but they don’t tangibly communicate Melissa’s enthusiasm about INFORMS, or Terry’s satisfaction with the association’s new direction or, for that matter, incoming President Anne Robinson’s commitment to marrying INFORMS with the burgeoning analytics movement unless you see them explain it for themselves.
INFORMS TV has transformed committed operations researchers and INFORMS leaders into more than a name on the page for the 4,000 plus operations researchers who attended this annual meeting. At the conclusion of the conference, we’ll share these films with the rest of INFORMS membership via the INFORMS Youtube channel.
The image on the screen has, of course, been virtual. But for many of us, it’s a step closer to tangible.

by Michele Fisher

As I left the Exhibition hall at lunch today, I observed the kick-off for the interactive session. The script for the announcement could have been used for speed dating. Five minutes only with the judges and rotate on the bell. Ready set go!

I stayed to watch a few rounds. I was impressed the concise summaries that were given of some pretty complex work. The talent to explain your analysis work to people from a range of backgrounds is an important analytical soft skill. One that can be honed with practice like that the poster session provides. It is important to be ready with an executive summary  of your work for the next time you encounter top management at a meeting, social event or even in the elevator.

by Steve Sashihara

Have you heard that INFORMS is launching a new program, where participants will receive the designation of Certified Analytics Professional (CAP)?

The scope / body of knowledge is briefly summarized in the following table:

Domain Description Weight
I Business Problem (Question) Framing 15%
II Analytics Problem Framing 17%
III Data 22%
IV Methodology (Approach) Selection 15%
V Model Building 6%
VI Deployment 9%
VII Lifecycle Management 6%

 

Eligibility Criteria:

  • BA/BS or MA/MS degree
  • At least five years of analytics work-related experience for BA/BS holder in a related area
  • At least three years of analytics work-related experience for MA/MS (or higher) holder in a related area
  • At least seven years of analytics work-related experience for BA/BS (or higher) holder in an unrelated area
  • Verification of soft skills/provision of business value by employer

Testing

For candidates who can satisfy the above eligibility criteria, CAP will be awarded upon successful completion of an exam. The very first exams will be administered at the upcoming INFORMS Conference on Business Analytics & O.R. in San Antonio TX April 7-9 2013.

If you’re interested in hearing more, the INFORMS Certification Task Force will make a short presentation and answer questions tonight (Tuesday) from 6:15-7:15PM in Hyatt — Regency A.

Here’s a tip: I think there might be nice to be one the very first to get this certification and use it as a differentiator.  It might help you stand out to potential employers.

 

by Michael Gorman

If you get a chance, you have to come see some of these presentations today – Tuesday ABCD — CC West, 212 C.

Last year’s first and second place competitors present at the 4:30 session….

The quality is superb!

 

by Michael Gorman

So much to do, so little time to do it!

From breakfast at 7 to desert at 9PM – trying to fit it all in.

Great things – the plenaries were fantastic… so were the improptu gatherings..

Bad things — walking from hotel to the North conference center in the blazing sun!

by Tallys Yunes

It was a very good Monday, ladies and gentlemen. I know I’m a bit (a lot) late with this post, but after enjoying the Traveling Salesman movie yesterday night, I was too tired to blog about my Monday activities. But here it is; better late than never!

I really enjoyed attending MA22: Generic Branch-and-Price Solvers. I didn’t know there was so much activity in that direction. The good thing is that the community of developers can learn and benefit from each other. On the other hand, as someone who is planning to use one of these packages in the near future, it will take me a while to sort through all the packages and pick the one that suits me best. So here goes a suggestion to the developers of GCG, DIP/DipPy, SAS Decomposition Solver, BaPCod (and others): would it be possible for you to build a comparison table? One row for every feature, one column for every package, and a checkmark in cell (i,j) of this table if package j implements feature i (with added footnotes to clarify technicalities when appropriate). I believe this would be very helpful.

BTW: You gotta love Ted Ralphs‘s choices of acronyms. “DIP and CHiPPS” was a clever one! I already know what my wife will say when she reads this: “You’re so food motivated…” I know! Speaking of food, after having a tough time eating a decent lunch on Sunday, I changed my strategy: have a large breakfast like the one on the left, and a quick small lunch.

;

;

;

;

;

Then came Bill Pulleyblank’s plenary. George Nemhauser introduced Bill and showed two interesting photos. The first was this one, of a young Pulleyblank:

After which George said (approximate quote): “Look at those eyes…you can tell he was already thinking about graphs and matchings.” I have to agree. You can tell the intellectual curiosity is there. The next photo, which I don’t have, is of Bill acting in a movie called “Wolf Boy”. I had no idea Bill was an actor, but it quickly became clear he’s very talented in front of an audience. His presentation had the perfect combination of interesting information and witty jokes.

Although most of us know what Erdos numbers are, not many people know about the Bacon number. It works just like the Erdos number, but rather than indicating how many co-authors one is away from Erdos in terms of published papers, it indicates how many co-actors one is away from Kevin Bacon in terms of movie appearances. By now you know where I’m going. The big question is: what is Bill Pulleyblank’s Erdos-Bacon number? (the sum of the two numbers). By using this site, I discovered that Bill’s Bacon number is at most 2. And this site tells me Bill’s Erdos number is 2. Therefore his Erdos-Bacon number is at most 4! Amazing! I want a finite Bacon number. Suggestion to INFORMS: can you please make a movie in which Bill Pulleyblank is the main actor and everyone else has a non-speaking part?

My next visit was to session MB29: the panel discussion on Bringing OR into the 21st Century with Social Networking and Web 2.0 Tools. If you’ve watched my video, this is the session I refer to in the second segment. I really enjoyed this session too. There were some very interesting tips, tricks, and perspectives from a number of different bloggers on their use of Twitter, WordPress, OR-Exchange, Facebook, LinkedIn, and more, with pros, cons, do’s and don’ts. Bjarni videotaped the session, so, if you missed it, he’s probably going to make the video available online somewhere. If you’re already part of this online world, or are thinking of joining it, this is a very useful resource. Interestingly, after the session, everyone started trying to guilt me into organizing next year’s edition of it. I’m not sure why.

This post is getting too long, so let me wrap it up. I also attended MC22, Tools and Techniques for Modeling with Python, which I’ve been considering getting into. It looks like Python makes a lot of things much more effortless than C++. I just have to gather some courage and take the plunge. Last but not least, I checked out MD17, Business Analytics Curriculum in Higher Education. I’m trying to put together an elective MBA class on Data Mining, which could perhaps be the first step of a broader analytics curriculum at my school. It was interesting to see what other schools are doing and what their curricula consist of. I now have some interesting pointers that I can further investigate after I return to the sunshine state (though there’s plenty of sunshine here as well).

To conclude, I want to leave you with two interesting/thought-provoking slides from Bill Pulleyblank’s plenary. Enjoy!

by John Angelis

INFORMS occasionally occurs at the same time as a big news event. For example, you may recall seeing researchers glued to the TV as their favorite football team played. This year, INFORMS takes place at the same time as the second American presidential debate, happening tonight.

Fitting operations to news events can be tricky, but can lead to an increased appreciation for the relevance of operations from non-operations people. (I should add that it also challenges the operations researcher to be as objective as possible about occasionally highly controversial topics). I would be interested in reading how you have been able to apply operations to news events in near-real time.

With full appreciation for the potential foolishness of this exercise, here are some possible applications of operations to tonight’s debate. For example, one could use game theory to explore the optimal personality strategy for a candidate in a debate. Given the different payoffs available for being poised versus being passionate, and the possible choices of the rival candidate, what is the candidate’s optimal strategy? And how does that strategy change given pre-debate signalling by both candidates as to which strategy they will select?

Or, consider that a candidate once rescued a basket of kittens from a burning pet shelter and wants to remind voters of this heroic fact. What is the optimal time for the candidate to mention this fact to impress the maximum number of undecided voters? Too early, and it may be forgotten: too late, and fewer voters may be watching.

Please do avoid the temptation to be partisan in the comments, but I am interested in reading how your favorite operations methodology may apply to tonight’s debate. I’d also like to read your favorite application of operations to a news event.

by Guillaume Roels

Today, I went on a tour of the Amazon fulfillment center Phoenix 6. Seeing the operations of one of the darlings of Wall Street was very impressive. As many would expect, speed is the motto of company: Every trip of packer is counted in seconds. Scale is also mind-boggling. The facility is huge and contains millions of SKUs. (Due to an NDA, I would refrain myself from giving you precise numbers, but believe me, they are very impressive). Another surprise, at least to people not familiar with their operations, is the labor-intensity of the picking and packing process. But above all, was the kaizen spirit and the dedication to perfect execution. In many respects, we, researchers who spend so much attention to details and preach for greater use of analytics, have found a master. Thank you again for making this visit possible, which nicely complemented the conference!

by Ken Chelst

On Sunday afternoon, Chairs of IE Depts warmly received a Project MINDSET update. A number of chairs are interested in using the material to develop a short math camp for highly motivated high school students. A handful have committed to using their schools high school relationships to launch Project MINDSET programs. A Partial listed of interested schools includes: U Texas El Paso, U of Tenn, U of Delaware, Texas A&M, U of Arizona, Cal Poly, U of Louisville, Harvey Mudd College

by Paul Rubin

On my flight into Phoenix, the first officer announced the inflight movie: Safetey Not Guaranteed. He had the giggles through the rest of his announcements. I’m not sure why being a panelist in the social media session reminded me of that. At one point, panelist Mike (@miketrick) Trick mentioned that searches for “zombies and operations research” drove traffic to Laura (@lamclay) McLay’s blog. (Laura has posted about zombies more than once.) Mike, arguably the most visible person from the OR community on social media, is also now an administrator. I was sorely tempted to ask whether searches for “vampires and operations research” pulled up his LinkedIn page. :-)

The MA22 session had a lot of good information about branch-price-and-cut solvers … but I would have had to sit through back to back sessions in the Military Applications track to see that many acronyms in one place. A couple of the presenters had slides featuring the cover of a (hopefully fake) book titled “Decomposition for Dummies”. I don’t think this was another zombie reference, but I can’t be certain.

In MD08, there were three presentations relating to Managing Operations in Non-profit Food Distribution, with methodologies ranging from algebra to neural networks to mixed integer programs. I’m a fan of MIP models, but it was refreshing to be reminded that basic algebraic models, correctly analyzed, can produce meaningful managerial insights.

by Laura McLay

INFORMS is putting together a series of videos about operations research. They have a YouTube channel here where they are uploading the latest and greatest. Here is one of the videos where INFORMS President Terry Harrison talks about the future of INFORMS:

by Laura McLay

I attended the WORMS (the Forum for Women in OR/MS) business meeting on Monday night and was extremely pleased with the turnout among students.  The issue of INFORMS membership came  up in the meeting, and I was genuinely surprised by a common misconception. It turns out that several student INFORMS members did not know that they had to separately join the WORMS forum. I am sure that this is not a WORMS issue – other student INFORMS members may not be registered for the Sections/Societies/Fora that you care about.

Students: it’s important to join Sections/Societies/Fora. When you join INFORMS, you get a lot of emails from INFORMS (a lot!!). You do not, however, get all of the INFORMS news that is relevant to you. Various INFORMS groups handle paper competitions, circulate job opportunities, put you in touch with people in your area. It’s important for your career to be part of the right conversations. The students who were interested in taking part of WORMS activities were not taking part in the conversations that happen on the WORMS list serve, and as a result, they missed out.

A lot of important business has been discussed at the INFORMS Annual Meeting. The conversations will continue after the annual meeting on various community listservs. Be a part of it.

If you have not yet registered for 2013, please do so online: http://www.informs.org/Membership/Renew-Your-Membership and, check  the various sections/societies/fora that are interesting to you.

If you have already registered and forgot to sign up for the various groups that are important to you, it is easy to remedy with one of three easy options.

  1. Join an INFORMS community here: http://www.informs.org/Membership/Join-INFORMS-and-or-INFORMS-Communities
  2. Renew by phone by calling INFORMS member services (800-446-3676).
  3. Renew by mail or fax using the form [pdf]: http://www.informs.org/content/download/261802/2462392/file/2013%20INFORMS%20Membership%20Application.pdf
by Laura McLay

The INFORMS Section on Public Programs, Services, and Needs awarded its best paper award at the SPPSN business meeting on Monday. This is an excellent opportunity to highlight papers that are used to apply operations research methodologies to areas that can make a difference. The application areas include elections, air traffic management, pollution, public defibrillation, and drugs in sub-saharan Africa.

  • Jeremie Gaillen (London Business School), Zachary Leung (Massachusetts Institute of Technology), Prashant Yadav (University of Michigan): “Rationality and Transparency in the Distribution of Essential Drugs in Sub-Saharan Africa: Analysis and Design of an Inventory Control System for Zambia” [Winner!!]
  • Krishnan Anand and Francois Giraud-Carrier (University of Utah): “Pollution Regulation and Production”
  • Luyi Gui, Atalay Atasu, Ozlem Ergun, L. Beril Toktay (Georgia Institute of Technology): “Fair and Efficient Implementation of Collective Extended Producer Responsibility Legislation”
  • Prem Swaroop, Michael Ball (University of Maryland): A Consensus-Building Mechanism for Setting Service Expectations in Air Traffic Flow Management”
  • Douglas King, Sheldon Jacobson (University of Illinois at Urbana-Champaign), Edward Sewell (Southern Illinois University): The Geo-Graph in Practice: Creating United States Congressional Districts from Census Blocks”
  • Timothy Chan, Derya Demirtas, Roy Kwon (University of Toronto): “Optimizing the Deployment of Public Access Defibrillators”

by Walt DeGrange
Disclaimer:I had lunch at Five Guys and ate a giant cheeseburger before this session. Since 80% of all of my body’s blood was in my stomach I apologize up front for discontinuities in the report.
Conference room was mid-sized. A very well attended session. Fortunately the AC was working well.

Repurposing the Broadband Spectrum

More broadband and less TV Stations. Sounds good to me! Who doesn’t want to send blog posts from INFORMS Conferences faster. Evidently getting TV stations to give up their bandwidth and move to other channels voluntarily is not as easy as it sounds. Use OR,  optimization and game theory to build a model to help. FCC is looking for assistance in this area. Nice!

Strategic Prioritization of Airline Delays
If you are flying ground-holding delays are not your friend. And since most of us flew to the conference then how could this presentation not be of interest. Unfortunately the briefed algorithm only used two airlines and was very difficult to scale up so better load up you kindle with some good books for the flight home just in case a ground delay is in your future.
Optimizing Kidney Exchange with Transplant Chains
You may not need a new kidney today but when you do it is pretty serious stuff. Evidently there are a lot of nice people willing to give up one of their two kidneys. Clearly most folks don’t do this out of the goodness of their heart. Typically folks will want someone they know but cannot donate a kidney to directly to get a kidney in return. There are a few very generous folks that do just give away a kidney and that creates a chain. Building and running models that make these chains optimal is an area where OR is really making a difference.
Tuesdays random session: 48 TSL – Freight Trans. & Logistics 8:00-9:30AM
by Steve Sashihara

Last night’s Awards Ceremony represented a small peek into one wonderful aspect of Operations Research: we are a community.

The honorees ranged in age from current undergraduates to very senior researchers, teachers, and leaders in our field.

It was fun to see in one hour a nice cross section of people at all stages of their careers — from students publishing their very first forays into advancing the art and science of OR — to the men and women who have blazed the trails many of the rest of us are following.

I was especially touched that Uriel Rothblum’s son was able to accept the prestigious INFORMS Expository Writing Prize on behalf of his father, who sadly passed away this Spring after a long and outstanding career contributing to a wide variety of topical areas.

George Nemhauser, accepting the John von Neumann Theory Prize on behalf of himself and Laurence Wolsey, was recognized for a lifetime of cutting edge contributions.

Andres Weintraub, in accepting the INFORMS President’s Award, mentioned how happy he was to accept an award for something that he has always loved doing.

 

by John Angelis

“So what are you doing in graduate school?” my inquisitive friend asked me at a 2003 Christmas party. I narrowly avoided blurting out “I don’t know!” before replying “I’m studying Operations Research,” expecting that the foreboding sound of “Research” would be enough to forestall any questions, as usual. “And what’s that all about, is it medical? Are you doing surgeries?” my friend said with a hint of mockery. “Well, it’s about inventory, and shipping, and…” I stammered. “So you’re going to work for UPS, right?” she said triumphantly. “You’re going to grad school so you can drive a UPS truck?”

 

Although at times I’m sure we all envy the bliss of the open road and an all-brown work outfit, that was hardly the impression I wished to leave with my friend! Since then, I’d like to think I’ve improved my opportunity to communicate about Operations. But there still have been plenty of opportunities for mistakes to occur. One of my favorite Operations “Losses in Translation” is overhearing a chat between a grad student and a tourist at a past conference. He was reading “The Black Swan” book by Nassim Nicholas Taleb, and she, wishing to make conversation, asked him how he enjoyed the movie.

As we are gathered here for the conference, I’d love to hear your anecdotes about what happened when you first tried to tell your loved ones about Operations. Please leave your comments below.

by Laura McLay

I had the pleasure of serving on the Doing Good with Good OR (DGWGOR) committee that judged the student paper competition on putting OR models and analysis into practice. The sessions occurred on Sunday (SB08 and SC08). The finalists are:

  • The Effect of Budgetary Restrictions on Breast Cancer Diagnostic Decisions by Mehmet Ayvaci, Oguzhan Alagoz, and Elizabeth Burnside
  • Dynamic Monitoring of Chronic Disease by Jonathan Helm and Gregg Schell
  • Optimal Distribution of Medical Backpacks and Health Surveillance Assistants in Malawi by Amber Kunkel, Elizabeth van Itallie, and Duo Wu
  • Resource-based Patient Prioritization in Mass-casualty Incidents by Alex Mills, Nilay Argon, and Serhan Ziya
  • Improving Patient Access to Surgical Care at the Juravinski Hospital Through Informed Decision Making by Daphne Sniekers
  • Using Simulation Methods to Guide Decision Making of the Kentucky Cabinet for Health and Family Services by Russell Harpring

The papers represent application areas from healthcare to social services to humanitarian work.

I’d like to briefly advertise the opportunities available in healthcare OR. The sheer number of finalists that published in the health care application area is evidence of how much healthcare OR research is being performed these days… and it’s growing. The papers, however, varied from screening breast cancer at the patient level, monitoring disease (e.g., glaucoma) progress at the patient level, and clinical improvements by improving how surgeries are scheduled in operating room. The opportunities in healthcare are varied and numerous.

If healthcare isn’t your thing, there are plenty of opportunities to do good with good OR in other areas.  Humanitarian projects, for example, are often not able to fund operations research projects, and as a result, fewer people study humanitarian applications. But OR can make more of an impact in these  emerging areas.

All of the projects were put into practice or have a path to be used either in policy or to obtain funding for supporting OR in low-resource settings. All of the presentations were excellent. It was clear that all of the students learned about “doing good” as much as they learned about “good OR” — they were terrific at answering questions about learning about the messiness of real-world applications and the excitement about getting “buy in.”

Kudos again to the students who completed these projects. I hope to see excellent student projects next year.

 

 

by Thiago Serra

image

Beyond the gifts mentioned at another blog post yesterday, Sulum Optimization innovated this year with a sponsored ribbon. If you are looking for more ribbons, come to talk with Bo Jensen. You can also end you finding a solver that is perfect for yours needs!

by Dawn Strickland

1. People are not staring at your chest, they are trying to see where you go to school. It only gets awkward when the person looking is actually from the same place but just didn’t recognize you.

2. Turn off your cell phone ringer when you go into a session. Don’t be THAT guy.

3. Nobody understands an entire presentation, unless giving it themselves. (Even then it is questionable.) If you can keep up with the first couple of slides, you’re doing fine. The person asking all the questions at the end is probably the presenter’s thesis adviser.

4. It’s okay to skip a session or two and check your email. (Or write a blog post.) Everyone does it.

5. Go to sessions and presentations given by other students. Also, go to the student reception tonight!  In general, go to any reception/lunch/gathering to which you are invited. It is at these smaller events that you will feel more comfortable, make more connections, and get free food.

by Allen Butler

There are lots of good talks scheduled for this meeting, but it’s sometimes hard to know which talks have good titles/abstracts and which ones also have good content. I know one track that will provide both high quality content and do so with an engaging presentation: the Wagner Prize track. I know this because I am the chair of the Wagner Prize Committee and I’ve read the papers describing the work that these top notch OR practitioners are doing. The topics are wide ranging this year, including three in the Healthcare industry (improving efficiency, saving lives), one in the political arena (minimizing gerrymandering), one in inventory management under extreme uncertainty, and a Homeland Security application that uses game theory to improve port security. Come hear all six talks, choose who you think was the best, and then come hear the winner announced at the 10 am Keynote on Wednesday.
Wagner Prize Sessions: Monday, Track 23 (MA, MB, MC) – Center – West 212C
One final recommendation: If you like single malt scotch, visit Seamus McCaffrey’s Irish Pub. They must have close to 100 choices and their prices are reasonable – they seem even more reasonable after the first few drams.

by Michael Trick

The Nobel Prize committee has never quite taken to operations research.  If George Dantzig  was not awarded the prize, it is hard to see what our field has to do in order to be recognized.  But many recipients are well-known in our field and their research has drawn from, and inspired, research in operations research.  This year’s Economics award recognizes two  economists, Al Roth and Lloyd Shapley, who are very well known in our field, and have strong ties to INFORMS and its predecessor organizations.

Shapley was recognized by ORSA with the von Neumann Theory Prize in 1981.  Here is part of the citation:

Lloyd Shapley has dominated game theory for the thirty-seven years since von Neumann and Morgenstern published their path-breaking book, The Theory of Games and Economic Behavior. Shapley’s key ideas include his inventions of convex games; stochastic games; the “Shapley” value for cooperative games; and his development of the theory of the core, including his independent discovery of the famous Bondareva-Shapley Theorem about the non-emptiness of the core. He has made important contributions to network flow theory and to non-atomic game theory. His work on the core influenced the development of fixed-point and complementarity theory, and his work on stochastic games influenced the development of dynamic programming. His individual work and his joint research with Martin Shubik has helped build bridges between game theory, economics, political science, and practice. Roth received the Lanchester Prize

Roth received the Lanchester Prize in 1990 for the book with coauthor Mari Ida Sotomayor, Two-Sided Matching: a Study in Game Theoretic Modeling and Analysis (Cambridge University Press, 1990).  The citation read, in part:

n their book, Alvin Roth and Mari lda Oliveira Sotomayor use policy evaluation and empirical observation as a guide to deep mathematical analysis. They demonstrate in precise, insightful detail how game theory in general, and matching markets in particular evolved into fields that are grounded in strong theory and, at the same time, are quite relevant to real issues of practice. The theory of matching markets, to which the authors have been major contributors, originated with the famous 1962 Gale-Shapley paper, ‘College Admissions and the Stability of Marriage.

The Prize Page notes that Roth’s comments included the following:

When I received my PhD in Operations Research almost 20 years ago, game theory was as much at home in O.R. as in any other discipline. Today I sit in an economics department, and the situation is very different. Game theory has grown enormously and become the backbone of modern economic theory, but in operations research it seems to be studied and used relatively little. This is an enormous missed opportunity, and one of our hopes for the book is that it should help explain the nature of this opportunity.

I don’t know if Roth’s comments hold today.  Certainly, game theory provides a strong underpinning to what we do, particularly in areas like supply chain and marketing with multiple actors with different objectives. And that area has grown tremendously since Roth’s comments in 1990.  There are more than 200 papers at this conference that have some aspect of game theory in their abstract.  So our field is certainly active in this area.  But there is a tremendous amount of work in the economics literature also.

Congratulations to Roth and Shapley.  It might not be a pure “Operations Research Nobel” but it is pretty darn close.

Image credits: From the official Nobel Prize page.

by Guillaume Roels

Although there hasn’t been much discussion on the blog about the conference theme this year, besides John Angelis’s blog about visualizing Informatics, I believe many of us share the opinion that Informs is facing a great opportunity for making an impact on business analytics. Business analytics is there. (See for instance The Wall Street Journal  piece a few weeks ago on the importance of big data for hiring decisions in large call centers.) It is up to us, academics, to seize that opportunity.

This opportunity is especially important for those of us who are affiliated with business schools. In fact, a recent article from The Economist highlights the importance for MBA programs to offer courses in business analytics. Dimitris Bertsimas mentioned to me today that MIT Sloan students expressed a strong interest in this area, even (or especially) at the executive level. Of course, MIT students may be more analytically driven than most other MBA students. But that example at least challenges the assumption that students are not interested in quantitative methods.

Following the lead of MIT, proper implementation of such a course would require us to repackage our traditional modeling courses, perhaps putting less emphasis on our bread-and-butter mathematical programming models to give more room for data mining and data analysis, or even for problem framing (as has been done at Stanford).

More importantly, the focus would need to be on strategic decisions, despite the traditional focus of OR/OM on operational problems. And the emphasis would need to be on making money (isn’t that “the Goal” after all?) rather than on cutting costs. (In fact, the big success of Revenue Management in the last decade may be attributed to the fact that it speaks the right language.) As an example of this kind of strategic project, Jun Li and Serguei Netessine presented today a study in which they identify clusters of competitors in the hotel industry in Manhattan, based on clickstream data. Their research niftily identifies competitors of a particular hotelier by looking at which other hotels are considered by prospective customers before they book a hotel room, thereby giving precise information to hoteliers about who their actual competitors are.

Most innovations in operations management in the last 30 years originated from industry: Total Quality Management (Toyota), Business Process Reengineering (Hammer), Enterprise Resource Planning (SAP), Customer Relationship Management (Siebel, Salesforce.com). This Business Analytics could be the time for OR/OM academicians to make a large-scale impact on industry while reinforcing the importance of our profession in business schools. Would we let this opportunity go? Feel free to share your best practices for developing courses on that topic.

by Walt DeGrange
Here is a link back to my Random Session Post.
Unique secret location on the third floor of the Hyatt. Kind of like Shangri-La although I didn’t stay long enough to figure out if I stopped aging. Below is a picture of the courtyard outside the conference room.
Small cozy conference room. Four briefs so this session was action packed from the get-go.
  • Presentations were done on a German ThinkPad. Evidently the keyboard is a little different.
  • Alternate ways to perform Benders Decomposition. “I don’t know about you but I am always on the lookout for alternate ways to perform Benders Decomposition.” And my wife tells me that OR doesn’t produce any good cocktail party lines.
  • ZIMPL 3.3 – Open Source Modeling Language for solving MIP and NPL. The example code solves a Sudoku puzzle. Now that is useful! And of course the price is right.
  • APMontitor Modeling Language – Uses servers not client software. Accessed through MATLAB, Python, Web Interface. Used with applications that require resolving problems every 2 seconds. the fastest application required reoptimization every 0.1 seconds. Makes waiting hours for my master thesis model to run so ten years ago.
Monday’s random session is Track 62 – Auctions 1:30-3:00PM
by Paul Rubin

Sometimes the real take-aways from presentations are not the details of the talk, but some remark a presenter made, or a question someone raised … or maybe just observations of the audience.

  • In the SA12 session (integer programming), James Ostrowski opined that antisymmetry constraints might make it harder to find a feasible solution by killing off the “clones” (my term) of any solution, leaving fewer feasible solutions to trip over. The same argument could be made for optimal solutions being harder to find. On the other hand, antisymmetry constraints may reduce dithering high in the search tree. So, on net, are they good or bad?
  • The SB08 session was the first of two sessions with presentations by the finalists in the “Doing Good with Good OR” competition. All three presentations were excellent (not universally true of sessions featuring more seasoned presenters), and the work seemed to be truly useful. Sadly, a schedule conflict ruled out my seeing the other session.
  • Two of the presentations in SB08 dealt with the systems in which a diagnostician interprets a test (or tests), leading to an estimate of the severity of an illness, which in turn triggers decisions about further diagnosis, treatment or whatever. Neither mentioned (at least that I heard) the possibility of using a diagnostician’s history of diagnoses and actual outcomes to modify the conversion of their assessment into probabilities for disease severity or progression. In other words, if diagnostician A is a bit of an alarmist and diagnostician B is possible a bit too laid back, that might be captured and used in a corrective manner. After all, it’s the age of informatics, right? (At least until next year’s meeting.)
  • The third presenter mentioned the difficulty in getting accurate census/demographic data in some poorer countries. I recall reading somewhere that there is a surprisingly high (to me) rate of ownership and use of cell phones in many impoverished areas. Cellular service providers might be able to supply data on the number of distinct phones used in different parts of the country (the cell towers provide locations). If available, could that data be used to sharpen estimates of populations by district or region?
  • I found it a bit odd that I personally supplied nearly all the gray hair in the audience. These talks were both inspirational and examples of how students can be engaged in important OR projects. That engagement will typically need support and encouragement from faculty. Are junior faculty more receptive to innovations like this than senior faculty?
  • SC16 was a tutorial on OR in humanitarian logistics. I attended a similar talk last year, but the content of this year’s talk was fresh (to me). The session was very informative and, for me, enjoyable. The audience hovered around 40, which is decent but less than my prior expectation given the social significance of the work. This may mean that previous efforts to get the word out have already exposed a large portion of the membership to it, or it may mean that future sessions need to be advertised more aggressively. It could of course just be a coincidence. (Was somebody giving out free food in another session? Tell me I didn’t miss free food!) It may also be that the scope of engagement in humanitarian logistics problems, and the need to work with actual end users, turn people off to the topic. I hope not, although I understand the concern about how much effort it would take to get started.
  • One of the issues that came up in SC16 was the need to get from models suitable for academic papers to simple but functional decision support tools (software) that NGOs can use to put those models to work. The academic system tends to skew rewards toward publishing papers and not toward implementation. (As a sidebar, I’m involved with the COIN-OR foundation, which deals with open source projects. We see a similar pattern there. Project leaders can be rewarded for writing and publishing code. Documenting and maintaining it is neither fun nor intellectually stimulating nor, typically, rewarded.) Getting back on topic, I wonder if there are opportunities to pair OR people (faculty, maybe retirees) with non-OR socially conscious programmers, the former explaining how the computations need to be done and the latter hacking code to implement the models?
by Tallys Yunes
  • Because cancer tumors typically are not well vascularized, chemotherapy can sometimes be ineffective (not enough blood vessels to take the drugs where they need to be). One idea is to load magnetic nanoparticles with drugs and use two body-sized magnets on each side of the patient to move these particles (and therefore the drugs) inside the patient’s body. Some of the research questions are: what should be the direction and duration of the magnetic fields?
  • Some companies have created web sites on which they display products they’re considering making (but aren’t making yet) and allow customers to vote for them (e.g. clothing and furniture). Some of the research questions are: how long should voting be kept open? And should the company offer monetary incentives (e.g. discounts) to elicit more votes?
  • On some web sites where those large banner ads appear at the top of the page, companies run real-time auctions behind the scenes to determine which advertiser will earn the right to show their ad. It all happens so fast that four such auctions can be conducted during the time it takes you to blink.
  • Robert Vanderbei’s work on high-contrast imaging for planet finding is AWESOME! How to design filters to be placed in front of telescopes so that stars (whose light makes them 10 billion times brighter than the planets close to them) don’t get in the way of our seeing the planets? NASA is likely to use his work in their next generation telescope!
  • One of the most challenging problems in humanitarian logistics is the last-mile delivery of goods. In addition, although a lot of research has been conducted on preparedness and response to disasters, the post-response recovery phase has received a lot less attention (e.g. debris management, restoration of infrastructure).
  • There is now a proof that every LP formulation of the TSP (and a few other problems) must have super-polynomial size. (No need to wonder about a bunch of “proofs” of P=NP any more.)

Not nearly as cool but relevant:

  • It’s very rewarding when people approach you and say they read your blog on a regular basis and had fun watching your goofy video. Thank you all for your support and encouragement.
  • The chairs in room C-West 207 are *very* comfortable.
  • The water in my hotel room isn’t very good at rinsing soap.
  • One hour for lunch isn’t nearly enough if you don’t know where you’re eating in advance, especially because long lines form quickly. Add vegetarianism to that equation for an extra layer of fun.
by Michael Trick

Open source optimization has been a big trend in operations research.  COIN-OR provides coordination for a number of operations research projects.  This evening, they announced the winner of the COIN-OR Cup.  The judges have given the award to a group of projects that don’t do optimization directly, but provide software for a necessary service:  interfacing!

After careful consideration of many excellent submissions, and much difficult deliberation, the COIN-OR Cup Judges have agreed on a 2012 COIN-OR Cup winner. The 2012 INFORMS COIN-OR Cup is awarded to a joint nomination of the managers and friends of three COIN-OR interfacing projects:

Bjarni Kristjansson for CoinMP,
Marcel Hunting and Marcel Roelofs for AIMMSlinks, and
Michael Bussieck, Steven Dirkse, and Stefan Vigerske for GAMSlinks.

In their deliberations, the committee members praised this team for the combined contributions they have made over many years to facilitate increased access to the COIN-OR solvers.

Bjarni Kristjansson has been developing and supporting CoinMP for 7 years now. COIN-MP provides easy access to COIN-OR solvers, such as CLP and CBC, from a variety of programming languages, including Bjarni’s own MPL Modeling System. A significant milestone for COIN-OR and CoinMP was the recent adoption of CoinMP for use in OpenOffice.

The COIN-OR AIMMSlink project was created in 2009 to build on and improve earlier work to integrate COIN-OR solvers with the AIMMS modelling system. Marcel Hunting and Marcel Roelofs both contributed to interfacing IPOPT and CBC with AIMMS. From 2010, these solvers have been distributed with the AIMMS software, making them available to all AIMMS users.

GAMS support for COIN-OR software dates back to Michael Bussieck’s work in 2004 with CBC. This work has continued and been extended so that GAMMS now also supports OSI, Ipopt, Bonmin, Couenne, SCIP, the Branch-Cut-Heuristic, and the Optimization Services project running on Linux, Solaris, Mac OS, and Windows. Members of the judging committee who have personally used GAMSlinks and found it very useful.

GAMS, AIMMS and the MPL Modeling System are significant players in the modeling arena and having interfaces to the COIN-OR solvers gives the COIN-OR solvers tremendous exposure.

During their development work, this team has also contributed to COIN-OR with solver improvements, bug reports and bug fixes. The judges were very pleased to learn that COIN-OR solvers are tested daily during GAMS and AIMMS regression runs on multiple platforms. It is thanks to the many years of high quality contributions from teams such as this that COIN-OR is so successful today.

The judges also wish to acknowledge the nomination by Andrea Lodi of Dr. Pierre Bonami as the main player in the development and contribution of Bonmin. It didn’t take much googling to see that Bonmin is making a huge impact in the optimisation community. Pierre’s most recent contribution is a nice piece of code that improves the performance of the Outer Approximation algorithm within Bonmin in the special case of Separable (Convex) MINLPs. The judges wish to formally recognise Pierre’s work by recording his contributions as being Highly Commended.

Coin-OR Cup Judges
Andrew Mason (OpenSolver)
Iain Dunning (OpenSolver)
Stu Mitchell (PuLP)
Kipp Martin (Optimization Services and COIN-OR Foundation member)

Congrats to the winners, and to the judges for recognizing this often unsung area!

by Dawn Strickland

Hello, INFORMS!

It’s that time of year where we awkwardly gaze at each other’s midsections, hoping to be discreet in glancing at the bright orange badges, pretending we remembered each other’s names and that we’re only looking to see if any job change has occurred over the past year. There are always a few people we can’t wait to see, a few people we recognize but can’t quite place, a few people we want to avoid at all costs, and then thousands more – 73 tracks worth! – that pass by in a blur (but not without glancing at your name badge). This is INFORMS – where the OR folks are in jeans, the MS people (and job seekers) are in suits, and everyone has matching tote bags. It’s great to be back!

by Ryan J. O’Neil

INFORMS 2012 may have just started, but in truth we’ll all be home in a few days. Besides the many hours spent in stimulating technical talks, networking with friends and colleagues, and overcoming jet lag, how are you going to remember the experience when you get home? Make it easy with this guide to the best conference swag currently available on the exhibition floor.

Elsevier is giving away these USB sticks to anyone who joins their mailing list. At first glance I thought “robot apocalypse” but at second glance they remind me of surgeons.

This fashionable bookmark is available from the Singapore University of Technology and Design. It’s made to spell SUTD while looking vaguely like a circuit board.

Gurobi Optimization has a free book on “Optimization for Dummies” that can help you explain what you do to your grandparents. Or your boss, for that matter. (When interviewed, one of the Gurobi folks said, “We’re giving away software! Doesn’t that count as swag?” No, guys. It does not.)

Finally, look sharp with these buttons from Tableau Software, proclaiming your elite data skills to the world.

Found any other great conference swag? Post a comment.

by Patrick Noonan

On Saturday, before the main conference began, the Hyatt was already becoming a busy place. As the site of this year’s Combined Colloquia – Future Academicians, Future Practitioners, and Teaching Effectiveness – the lobby and meeting rooms were the gathering spots for a diverse mix of participants and presenters.

As the Chair for this year’s Teaching Effectiveness Colloquium, I got my first look at what happens on those pre-meeting Saturdays, and it was a great experience.

In the Teaching Effectiveness Colloquium, our participants represented a wide range of INFORMS members. Some were just finishing their Doctorates, others were wrestling with their first real teaching responsibilities, still others were old hands just taking the opportunity to devote a day to reflecting on teaching and learning. Some were from engineering departments, many from business schools; some represented public institutions (including one service academy), others private ones. Collectively we were optimizers, schedulers, decision and risk analysts, statisticians, modelers, and nearly every other flavor of the INFORMS community.

All were there to hear from – and discuss with – a terrific team of experts in OR/MS education. Over the course of a very full 9-hour day, the participants learned a lot, of which we can present only a brief summary:

  • Donna Llewellyn (Georgia Tech) spoke about “what we know about teaching and learning.” The research on learning tells us a lot about what works and why, and provides important frameworks for thinking about sound educational practice. Considering a short list of well-understood factors goes a long way in making our course designs and interactions with students more effective.
  • Jill Hardin Wilson (Northwestern) provided more of this context for teaching by focusing on “mental models.” Students arrive already loaded with mental models – ways of thinking about and organizing their understanding of how the world works – and we need to work with those existing models, often challenging them, and occasionally disrupting them.
  • Peter Bell (The Ivey School, Univ. of W. Ontario) made the case for “bringing the real world into the classroom.” He offered a wide range of prescriptions for doing so, from pedagogical suggestions (e.g., activities, cases, projects) to institutional ones (e.g., rethinking how we label and position ourselves and our courses). He also shared some views on the meaning of the elusive “analytics.”
  • Jeff Camm (Univ. of Cincinnati) demonstrated his approach to “teaching exploratory optimization,” in which he challenges students to think about what questions and out-of-model issues a real world client would be thinking about. Students learned new modeling tricks, but also valuable practical perspectives.
  • Susan Martonosi (Harvey Mudd College) described how to incorporate field projects into a curriculum, reminding us of the enormous value of taking students “off road.” The complexities and complications of the real world reminded her of a challenging trek in Africa, inspiring her title, “Pretty cars don’t go here.” However, she surveyed a variety of less-daunting ways to start incorporating project experiences.
  • Finally, Patrick Noonan (yours truly, from Emory University) provided a quick overview of how to use cases and discussion leadership as another way to bring active learning into our classrooms. Teacher centered learning is effective for some of what we do, but it has cognitive, philosophical and practical limitations. There are a variety of ways to add cases to the mix and make learning more active, enjoyable, and effective.
  • We were fortunate to have such thoughtful facilitators throughout the day, but it was the participants who really made the day work. The questions and discussions we were able to elicit brought the material to life, and spread the learning around. (Side note: I’ve been teaching now for 20 years, and my office window sill has no shortage of shiny recognitions for, well, appearing to know what I’m doing – but I myself found myself taking lots of notes in every session. Even the oldest dogs can learn new tricks, especially when it comes to teaching and learning!)

    Many thanks to the entire group for devoting their day to sharing their ideas and experiences!

    by Ryan J. O’Neil

    For any of you students who are graduating soon, JPS had a breakfast session about finding jobs in academia and industry. In case you didn’t get to attend, here are a few notes.

    When looking for a job in academia:

    • The success of your research is to a certain extent your own issue, but the quality of your teaching can have a big impact on the rest of a department.
    • All the senior faculty members have already taught the core courses in their departments. Part of hiring a junior faculty member (you) is so they don’t have to anymore. This is the way of things.
    • The current faculty members probably won’t see you as a threat to their research work, but may be concerned if you have clear ideas about what you want to teach. Be willing and geniunely excited to teach anything the department needs.

    When looking for a job in industry:

    • LinkedIn is your friend. A lot of employers search for and directly contact good candidates these days instead of the other way around.
    • But try and do all your LinkedIn updates at once as it has a tendency to spam your contacts.
    • One of the most important skills you can have is the ability to handle data. In school you have likely solved problems where the technological coefficients are given to you. But in industry, you’re going to have to get them yourself. Learn how to use a database, how to use Excel, etc.
    • Learn as much as you can about the company interviewing you. If they’re publicly traded, look at their earnings statements.

    When looking for either:

    • Talk about why you want to work at the place you’re interviewing, not why you want to leave the place you’re at.
    • The only correct response to an interviewer asking a negative question about someone else’s work is, “I’m afraid I’m not familiar with it.”
    • Dress appropriately. A lot of how you’re initially perceived relates to appearance. The old cliche, “you never get a second chance to make a first impression,” has a lot of merit.
    • Remember, the hiring process is stressful for intervewers too! Sometimes they interview an enormous number of candidates over long periods of time, only to be rejected by their final picks.

    Please add anything I might have missed, along with your own advice!

    by Michele Fisher
    When I opened up my registration envelope – there it was – my 25+ year ribbon. That ribbon marks 25 years of a career in Operations Research. I was very lucky to be given the opportunity to attend an ORSA/TIMS Conference in Denver Colorado and present work on my very first project. By the end of the conference I had signed up as a member (at a discount- of course). I have maintained that membership as my career progressed to tackle different problem sets in different parts of the world. INFORMS resources and contacts were always there to support me. My 25+ year ribbon marks a long and rewarding relationship with INFORMS.  I will be wearing it with pride.
    I also plan to welcome the first timers I encounter and hope that this conference will mark the start of a rewarding relationship with INFORMS. I encourage you to do the same.

     

     

    by Ken Chelst

    I am on my way to present to department chairs an update of Project MINDSET.
    Goal: Seek more apostles to reach out to HS
    Project MINDSET – 4th year high school math course (with textbook) based on OR. It was developed with $3.3 million NSF funding. Course is two semesters: 1) Deterministic and 2) Probabilistic
    Partner universities (Wayne State U., NC State, UNC Charlotte and Georgia Tech)
    4000+ high school students study curriculum in dozens of high schools in MI, NC, and GA
    Statewide roll out in progress in NC and GA
    Individual schools in NYC, Indianapolis, and San Diego suburb.

    by Walt DeGrange

    The INFORMS 2012 Annual Meeting for me is Tweeting, blogging, meetings, exhibits and all of those wonderful sessions! So now I look at the Master Track Schedule and plan my attack. This year I thought I would try something a little different and attend a totally random (I am using Excel so maybe not totally random) session. I am selecting the time frame and then using the RANDBETWEEN(1,73) Excel function to guide me. After attending the random session I will provide a review on the blog.  Haven’t figured out what I will do if there is “No Session” listed in the randomly selected event. Perhaps a nap…

    Today’s time frame is SD 4:30-6:00 and the lucky track is… 72! ICS-Advance in Opt. Mod. Languages & Systems here I come!

    by Guillaume Roels

    The light rail system is very convenient; I recommend that everyone give it a try. Although I had never been to Phoenix before, I adventurously followed the succinct (but very useful!) recommendations from the conference website to get from the airport to my hotel by light rail. And I have to say that I found the system very easy to use, clean, quick, and reliable. This is very surprising to me, given the city’s image of being “the world’s least sustainable city“; yet few American cities could pride themselves on having such a shiny and efficient public transportation system.

    Looking at ValleyMetro‘s website, I saw that ridership is still increasing (and is currently at +40,000 daily users on weekdays), and the network is still expanding; and I suppose that its complexity and congestion would also be increasing. Although sustainability was the theme of the conference a few years ago, there seem to remain plenty of opportunities for OR specialists in that area!

    Many thanks to the conference organizers for drawing our attention on this great public transportation system. This, together with the new mobile app that should reduce the need to order the printed copy of the program (although I still like better using the pdf version of the program), should hopefully reduce the environmental impact of this increasingly big conference.

    by Jim Cochran

    Sunday is generally an unusually hectic day for me at INFORMS conferences – there always seems to be several sessions I want to attend throughout the day, and I have to make some difficult choices (a really good problem to have, if you must have a problem!).  With 73 concurrent sessions at this conference, I am facing the same problem this year. One decision is easy – at 4:40 p.m. I will be in Regency Ballroom A in the Hyatt for the session, “Feature Articles from Recent Issues of INFORMS Transactions on Education (ITE) and Discussion of the ITE Review & Publication Process.” As the  Editor in Chief of ITE I may be biased, but I think this is going to be a terrific session. Several members of the ITE editorial board will speak about their roles with the journal, and several authors of  papers published in recent issues of ITE will discuss their work. This is a great opportunity to learn about the journal (including how to access classroom cases published by ITE – which are now in use in over 70 nations). Unfortunately (for me), the session, “United States Presidential Election Forecasting: Who Will Win the White House in 2012?” is scheduled opposite the session on ITE; this session should be fascinating, and I am sorry I will miss it.

    Two sessions sponsored by the SpORts section are scheduled for tomorrow, as are the two CPMS sponsored sessions featuring presentations by 2012 Edelman Award finalists (the winning team is also scheduled to reprise its presentation in a keynote session at 3:10 p.m.). Rich O’Lear’s plenary and George Poste’s keynote should both also be very interesting. And then the INFORM-ED (the INFORMS Education Forum) Member Reception and Business Meeting in West 105B of the Convention Center from 6:15-7:15 p.m. and the Awards Ceremony soon after, so this will be a busy (and fun) day.

    by John Angelis

    One of the most exciting trends in informatics is the increased intricacy and variety of informatics visualization. Much of the reading public may be unable to keep up with the mathematical results or charts that are produced by informatics. But as time goes on, graphic designers and software programmers are providing richer visual options for researchers to display their results.

    For example, take the sport of basketball. As a child, you may have collected basketball cards, which listed basic statistics such as points per game on the back. You may have mused what it meant that your favorite player scored “16.2” this season compared to “15.9” last season, but there was not much depth or context to the statistics provided. See, for example, this 1990 card:

    However, take a look at some of the graphics now being created to present basketball informatics:

    Shot Selection Heat Chart

    This heat chart by researcher Kirk Goldsberry shows where players tend to shoot from and how successful they are from each location. Or, on a simpler note, take this distribution of passing skills from sportsvu.com:

    These infographics contain the results of complicated, labor-intensive processes, yet most readers can easily interpret the basic meaning of the research. As a starter blog, I wanted to ask you, what infographics have you seen or produced that you are excited about? Please link them below or describe them.

    by Tallys Yunes

    Like most people by now, I’m getting ready to fly to Phoenix (bags packed, boarding pass printed). Due to busy work weeks leading up to this trip (including a midterm exam that’s taking place Saturday morning), I did not have a chance to go through the conference program in detail yet, so I’m still not sure where I’ll be on Sunday (except for my own presentation at 4:30pm, the Social Networking Reception at 6pm, and the Welcome Reception at 7:30pm). But this is what the flight is for, right? This and rehearsing my presentation: the slides are ready, but I don’t know how long it takes to go through them yet. Luckily, I’ve done this enough times that I’m not too far from the target 20 minutes.

    As for choosing presentations to attend, I’ve decided to try something different this year. I’ll step out of my comfort zone and attend at least one session that I’d normally not consider attending. Something completely unrelated to my usual choices and to my current research interests. In doing so, my goal is to open my mind to something new: a new sub-area of OR, a new problem, new ideas. Hopefully, most presenters will follow Paul Rubin’s advice, which will increase my chances of being able to follow what’s going on (since I’ll probably lack some background knowledge). I’ll let you know about the outcome of this adventure in a future blog post.

    May you all have a safe trip to Phoenix. I’m looking forward to changing from humid heat to dry heat.

    See you soon!

    by Thiago Serra

    The Constraint Programming (CP) community just had its annual conference in Quebec City, Canada. Taking advantage of their presence in North America right before the INFORMS Annual Meeting, a number of CP practitioners from varied parts of the world is now heading from Quebec City to Phoenix for the CP sessions organized by CMU Professor van Hoeve (and for a thermal shock as well, since we were used to temperatures between 2 a 11 degrees Celsius during the CP conference, and we even saw snow last Thursday).

    Anyway, what is CP?

    CP may interest those who enjoy solving combinatorial problems and scheduling problems in particular. It is a technique that attempts to exploit the structure of combinatorial problems by means of a comprehensive set of global constraints that represent special relations that would be much more difficult to state with standard mathematical programming models. Beyond the compact modeling, what really makes the strength of CP is the propagation algorithms associated with those global constraints. They have specialized inference algorithms to rule out huge chunks of search space that do not contain any feasible solution. In some cases, it is even possible to prove that a given problem does not have a solution at all. Associated with those propagators, there are search mechanisms based on backtracking as well as on Large Neighborhood Search (LNS), in the case of more recent solvers. And even though optimality is hardly achieved with CP, there are many cases where it finds good solutions and other approaches would hardly find any, as is the case of many scheduling applications in the industry.

    What, when and where?

    Joint Session Optimization IP/ICS: Constraint Programming Methodology and Applications I
    Tuesday Oct 16, 11:00 – 12:30
    West 106 A

    Joint Session Optimization IP/ICS: Constraint Programming Methodology and Applications II
    Tuesday Oct 16, 13:30 – 15:00
    West 106 A


    ABOUT

    divider
    CONFERENCE NEWS

    divider