Michael Trick’s Operations Research Blog

Don Ratliff at IFORS

So what happened to my “live blogging” at IFORS 2008? Well, unfortunately I had an administrative role at the conference (nothing like the real work of Hans Ittmann and John Bartholdi, but a role none-the-less) and that took time. And I am always a believer in the “social networking” aspects (read: hanging out at the bar with friends), and that takes time. And I did go on one of the famous IFORS outings, where I got sucked at by an elephant. Finally, I spent 3 days at the end at a lodge with no internet connection (technically, there was an internet connection, but I pretended it didn’t exist). Put it all together, and I fell a bit behind. But here are some notes that I took along the way.

The Tuesday plenary speaker was my co-adviser from 20 years ago, Don Ratliff from Georgia Tech. Don has had a great career, both in academia and in business. He was editor of Operations Research for a while, and published a number of interesting papers. He also founded CAPS Logistics, which was later bought out by Baan in a rather confusing and lawsuit-laden muddle. The title of his talk was “The Role of Operations Research in Lean Supply Chains”. He began by drawing an analogy between lean production and lean supply chains. In his view, lean production involves three main components:

1. Eliminating waste. Waste can be in terms of inventory, or time, or anything else that is not needed to meet the ultimate end of production: producing the thing!
2. Synchronizing flow. Getting things to where they need to be at exactly the right time.
3. Continuous improvement. Production environments are not static, and there is the opportunity for improvement as processes change.

He suggested that there has been insufficient effort to adopt these precepts throughout the supply chain. Partially this is a result of the multiple actors within a supply chain. It is typically easier to say “do this” within a factory with a common boss; supply chains often have multiple people involved, including the customer, making it harder to do things like waste removal and synchronization. But there are significant savings possible, making things worthwhile.

I thought the most interesting point he made came under the area of “continuous improvement”. He pointed out that ERP (Enterprise Resource Planning, the SAPs and Manugistics of the world) have shut OR out. It is hard to improve part of the system, when that part is embedded in a larger whole. In essence, OR has missed out of ERP, so nothing much has changed in OR since the PC (which did have an effect on the OR world and thinking). But he identifies SOA (Service Oriented Architecture) as a great new opportunity for our field. Quoting from wikipedia:

Service-Oriented Architecture (SOA) is a software architecture where functionality is grouped around business processes and packaged as interoperable services.

It is through these services that OR can have its strongest effect. If you have a better routing system, you can embed that in the routing service of the SOA system, improving that aspect without affecting the rest of the system. This is exactly what is needed for continuous improvement, and exactly what OR is good for.

Within OR, we often don’t track IT concepts such as SOA or business intelligence, but we should: it can have a great effect on how our work is used in organizations.

I thought this was another very good plenary, and again the audience seemed quite engaged with in.

Opening Session at IFORS

After a somewhat rocky start (an electricity substation in Joburg failed, resulting in large traffic delays, so the student assistants were late so the rooms didn’t have computers on time), the IFORS conference officially opened with a fantastic opening session. Of course, I am biased, since I am on the IFORS board with responsibility for meetings. But I do think it went very well.

First, there was dancing and singing by a choir and drums which set the African mood. Then Elise del Rosario gave a terrific welcome, talking about the promise of Africa and the role operations research plays in achieving that promise. This was followed by a welcome by the Minister for Science and Techology, Mosibudi Mangena. Minister Mangena is one of us: he has a masters degree in applied math, and he gave a talk that suggested he understood operations research. As a colleague said, the talk also suggested a speech writer familiar with the Science of Better web site. The minister took the opportunity to tweak Hans Ittmann, Organizing Chair for the conference, for not offering enough operations research to his department. Hans promptly promised to work hard to embed OR throughout the South African government. Could people even 15 years ago imagine a South Africa where an esteemed black Cabinet Minister would tease a (white) operations research professor about the role operations research plays in the government? I will write later about how I feel holding a conference in South Africa at this time, but I loved the exchange.

One person who gets credit in South Africa for the amazing transformation over the last two decades was the opening speaker Clem Sunter. Clem is a Brit who has been in Africa for 35 years, working for the Anglo American Corporation and more recently heading up the Anglo American Chairman’s Fund, a corporate social responsibility fund.

Clem does “scenario planning” and has been described as Africa’s Alvin Toffler. In short, he is a futurist. In the 1980s, he developed two scenarios for South Africa. The “High Road” was a South Africa transformed by negotiation leading to political settlement (I do not need to remind people that South Africa in the 1980s was torn by the odious apartheid). The “Low Road” was confrontation leading to a wasteland. He talked about these scenarios to both FW de Klerk, leader of the National Party, and the imprisoned Nelson Mandela. Mandela and de Klerk shared the Nobel Prize for their combined work to end apartheid.

Clem opened his talk by suggesting scenario planning is like poetry, where operations research is like an essay. He uses possible futures to frame debates about where companies, countries, and societies are going. When speaking about South Africa, he clearly sees the problems: crime, HIV/AIDS, and a declining infrastructure for some. But he also understands the strengths: resources, tourism, and its role (like that of Dubai in the mid-east and Hong Kong in Asia) as a gateway to Africa. In his view, one key to success is the turnaround in Zimbabwe. He believes that President Mugabe is in an endgame, and once he is gone, tremendous investment will flow back into Zimbabwe, with much going through South Africa, and with much of the rebuilding being done by South Africans.

I must say I am not a great fan of futurists. They are either so vague they become like horoscopes in that you can read almost anything into what they say, or make so many “predictions” that almost any future will have been predicted by them. Clem seemed more thoughtful on the purpose of scenarios. He has scenarios not to predict the future but to frame the debate. It is in this approach that he sees poetry in scenario planning.

There was a terrific question right at the end of the talk (if anyone noted who asked the question, please let me know) about the role of operations research in scenario planning. We do scenarios all the time: we create stochastic models of the future and simulation then provides us with scenarios that we can optimize over. How does this relate to Clem’s use of scenarios. He said that there was a large difference, whereby his scenarios change conversation while technical scenario building has not had the same broader impact. And, he said, there is a Nobel Prize in the waiting for someone who combines the two. You can find out more about Clem and his thoughts at a website for him and his colleague Chantell Ilbury

I enjoyed the talk a lot, and learned a lot about South Africa and the challenges the country faces. And I was a little inspired to think about how the things I do can change conversations. I am sure there are differing opinions, since this was not a typical OR talk.

Unusually for an operations research conference, not a single person (out of 500 or more) left the plenary once Clem began speaking. That is the sign of a successful opening plenary!

Up tomorrow, back to traditional OR with one of my favorite people in the field (and my co-advisor 20 years ago), Don Ratliff from Georgia Tech on Lean Supply Chains.

In South Africa

I’ve arrived in South Africa (Sandton) and am ready for the IFORS conference. I’ll try to do some posting along the way so that all you poor people who couldn’t make it here will at least have a glimpse of what is going on. First up: a day of IFORS Executive Meetings! That should put my ratings through the roof.

Who Knows Where Operations Research Will Lead You?

One of the nice aspects of working in operations research is that you can end up working in practically any field. I know a lot about the United States Postal Service, Major League Baseball, auction design, voting systems, and many other areas because that is where my research and reading in operations research took me.

Compared to Ronald Johnson, however, I am hopelessly narrow in my skills and interests. Major General Johnson was, until recently, the number two engineer in the US Army, as reported in the New York Times (thanks to Barry List from INFORMS for the pointer). His responsibilities were described as follows:

Before retiring from the military, Johnson was the deputy commanding general of the Army Corps of Engineers, the second-highest-ranking engineer in the Army. He supervised $18 billion of reconstruction projects in Iraq from 2003 to 2004 and commanded the 130th Combat Engineer Brigade in Bosnia from 1996 to 1998.

Now, however, he has a new job: he was hired by the National Basketball Association to be in charge of their referees.

While Johnson readily acknowledges that he does not know anything about refereeing, he knows quite a bit about difficult rebuilding efforts.

Why was he able to make this sort of career change?

N.B.A. officials are highlighting Johnson’s management and analytical skills.

And where did he get those analytical skills? Operations research, of course.

Unquestionably, Johnson did not take the typical career path to the N.B.A.’s executive suites. The commissioner’s office has generally been populated by lawyers and basketball people. Johnson, a 1976 graduate of West Point, studied mathematics and mechanical engineering. He later earned a master’s degree in operations research and systems analysis from Georgia Tech’s School of Industrial Engineering, and a master’s degree in strategy from the Army’s School of Advanced Military Studies.

This is a great example of the flexibility analytical skills provide in one’s career.

Operations Management Blog

Gerard Cachon, who I overlapped with in New Zealand, along with Christian Terwiesch have started a blog on operations management issues.  They get to the heart of the issue in naming the blog “Matching Supply with Demand” which is about as good a tag line as I have seen for operations management (it is also the title of their book).    The blog has just five posts so far, but is very much in keeping with the goal of looking at current events from an OM perspective.  Check it out!

NL8 Solved!

Almost 10 years ago, Kelly Easton, George  Nemhauser and I created something called the Traveling Tournament Problem.  The name is George’s:  I wanted to call it the Problem where Teams Play a Round Robin and Travel and Stuff, but George’s name is catchier.  The problem came from work we had done with Major League Baseball in creating their schedule.  Rather than try to post 150 pages of requests and requirements (which MLB would never let us do anyway), we abstracted out the key aspects of the schedule.  Those were:

1. Travel:  a wish to keep the total travel down, and
2. Flow: a wish not to be home or away for more than 3 series consecutively.

I also added a “no repeat” constraint:  if A is at B, then B cannot be at A in the next time slot.

When we put this together, we also put together the website http://mat.tepper.cmu.edu/TOURN to keep track of results and that turned out to be a great idea.  By having one site with results and instances, it has been easy to keep track of “current records”.

When I first worked on this, my goal was to show how some of our methods that we use for MLB could handle problems of reasonable size for the Traveling Tournament Problem.  That turned out to be impossible to do.  The Traveling Tournament Problem is hard!  Even solving the eight team problem turned out to be nontrivial.  Six teams is doable with a bit of work.

Kelly Easton, in her dissertation, put a lot of effort into the TTP, and even solved the eight team problem.  She used a network of 20 computers over the course of a week to prove optimality.  Unfortunately, she did not include the “no repeat” constraint, so we didn’t have the result for exactly the right problem.  While we believed that adding the “no repeat” constraint wouldn’t affect things too much, Kelly graduated, began working (for my small sports scheduling business) and we never solved the real problem.

Despite lots and lots of people working on the TTP, proving the optimality of a known solution to NL8 with value 39721 has been elusive.  In fact, relatively little (but some, thanks to Melo, Ribeiro, and Urrutia) work has been done on improving lower bounds.

I was thrilled yesterday to get an email Stefan Irnich of Aachen who, together with his master student Ulrich Schrempp, claims to have proved the optimality of 39721.  But here is when it gets hard.  How do I know?  It is easy to check feasible solutions, but unless there is a simply calculated lower bound, it it hard to confirm optimality.   Fortunately Stefan sent the slides of his work, and it seems clear that the approach they are taking is one that would work, and would be stronger than what others have tried.  So I have no hesitation in proclaiming that NL8 is now solved!  I am sure there are easily dozens (OK, half-dozens … OK, a half-dozen) people who are excited about this, and I am one of them (and it is my blog).

On to NL10!

Data Visualization

I have always loved Data Visualization (well, always since my adviser John Bartholdi pointed me to Tufte’s classic “Visual Display of Quantitative Information”). I teach data mining here to our MBAs, and have wanted to include the topic, but never knew what to include. Thanks to Stephen Baker of Business Week and his pointer to Many Eyes, I think I am getting an idea. Many Eyes is an IBM site with a goal of making data visualization algorithms and data sets widely available. It is a fantastic place to spend a few hours. As an example of what you can do on the site, here is a tag cloud of my vita (the source is at http://mat.tepper.cmu.edu/trick/vita.pdf):

I think you can find a fair amount about me just by looking at that tag cloud, though I am a bit biased (most ink blots end up looking like me in my eyes). Perhaps even more than by reading through a 12 page vita (by the way, vita (or curriculum vitae) is supposed to mean “a short account of one’s career and qualifications prepared typically by an applicant for a position”. What ever happened to short? What is the name for the document where you put down every blessed thing you ever did in your academic career?)

The structure of Many Eyes is unusual: you don’t download computer software. Instead, you upload your data (which immediately becomes public, so don’t try this with your financial records) and work with it there. This means that Many Eyes is quickly collecting a huge amount of data (23,256 data sets so far) that it (and you and others) can work with. This “social networking” aspect is unexpected, but I would bet that some interesting results come from it.

Another fascinating site is Wordle, which also creates tag clouds, but does so in a more artistic way. Here is my vita in that form (a couple of versions). I think I will use it during my next salary review!

I think I will need a few more days to recover from my surgery before I can get any useful work done.

IFORS 50th

The International Federation of Operational Research Societies (IFORS) is holding a conference in a few weeks in South Africa. Part of the festivities is a celebration of the 50th anniversary of the organization. I am putting together a presentation on the history of IFORS and would very much welcome anything anyone has that might be relevant (including pictures from conferences, pictures of Presidents (pre-1995 or so: I think I can find Tom Magnanti on the web!) and so on. Let me (trick@cmu.edu) know if you have something! Please, please, please!

There is a wonderful indexed photo of the first international conference (Oxford 1957) available. It is too bad that conferences have grown so big that such pictures are now hardly ever taken. It is fascinating to see some of the “big names” in our field as young men, and sometimes women. Can you pick out Dantzig without checking the index? Hint: he is in the front row.

One striking aspect of the history is the variety of backgrounds of the early Presidents of IFORS. Sir Charles Goodeve was the first president of IFORS (I have a family connection with Sir Charles). He had done tremendous work in OR during World War II. The next President, Philip Morse, also did significant work during the war and went on to found the Bookhaven National Laboratory, among other institutes. The third president, Marcel Boiteux, led French Electricity and was a leading proponent of nuclear energy for that country. After Charles Salzmann finished Boiteux’ term (I know little about Salzmann), Alec Lee, a manager at Rolls-Royce became President. That is two physicists, an economist, and an automobile executive as the first presidents! Many more recent presidents (Pierskalla, Bell, Weintraub, Toth, Magnanti) are mainstream academics, though the current president (Elise del Rosario) made her mark with the San Miguel corporation, a food, beverage, and packaging company in the Philippines and South East Asia.

It has been interesting to look into the history of IFORS. For a number of reasons (primarily because its members are societies, not individuals), it is less well known than groups like INFORMS, but I am glad to be part of it. More about IFORS in the coming weeks leading up to South Africa.

First computer, now me on the mend

Now that I have the “new” computer working, I have to spend a couple of days on the mend myself:  I had surgery for a hernia on Wednesday.  My plan was to to a “live blog” on the efficiency of hospital operating procedures.  I guess I can tell you that general anesthesia works pretty well, since I have absolutely no memory between “Just slide onto the table here” and “How are you feeling?  Ready to go?”.  So give me a couple of days then I will be ready to take on the OR world again, with upcoming conferences in South Africa and New York.

Perhaps the New Machine is Working

If you see this, rather than the previous message “This blog is moving”, this means the blog has moved, and I should be active again!  Please let me (trick@cmu.edu) know of any problems.