Test Gurobi yourself!

In keeping with an unusually awesome collection of comments, Daniel Fylstra of Frontline Systems wrote regarding the new linear/integer programming software from Gurobi (apologies for the self-linking, but, hey!, it’s my blog!):

Although Gurobi Optimization plans to offer the Gurobi Solver directly (with its own APIs and tool set) starting in April, it’s available in a full commercial release this month (February) as embedded in several modeling systems (AIMMS, Frontline, GAMS, MPL). We have it available for download right now (15 day free trial with no other restrictions) at http://www.solver.com/gurobi. You can run the Mittleman benchmarks yourself, or solve your own models in MPS, LP and OSIL format using a little program GurobiEval that we include. You can create models from scratch, and solve them with Gurobi, either in Excel with our modeling system, Risk Solver Platform, or in C, C++, C#, VB6, VB.NET, Java or MATLAB using our Solver Platform SDK.

This is excellent news, and I look forward to experimenting with the new software.

If this level of comments keeps up, I’ll have to initiate a “quote of the week”!

Keep the Operations Research Name!

A few years ago, INFORMS spent significant amounts of money on its “Science of Better” campaign in an effort to brand the phrase “Operations Research”.  This was not an easy or uncontroversial decision.  Even choosing “Operations Research” was controversial.  Practically everyone agrees that it is a lousy name for a field:  it is uninformative and easily confused with many other fields.  But it is the historical name for the field, and trying to rebrand our field with a new name looked (and looks) hopeless.  So “Operations Research” it is, and the Science of Better campaign tried to get people to associate the phrase with success, competitive advantage, and all sorts of other good things.  People like Tom Cook, Irv Lustig, and many others put in tremendous efforts to get this campaign going.

By its nature, an advertising campaign for an idea is a hard thing to do, particularly for a society  of just 12,000 or so.  Spending a million or two was possible, but not tens or hundreds of millions.  That 30 second ad at the Super Bowl was right out.  The campaign did lots of interesting things, some of which (like the website and the enhanced Edelman Award) continue, but it is hard to find a meter that moved because of the campaign.  Despite that, I am very happy we had the campaign:  it reminded a lot of us as to the value of our field, and let others know of that value.

However, just as cities like Pittsburgh get tremendous advertising by being backdrops for national sporting event telecasts, our field gets its best advertising just in how departments and schools get mentioned.  In the New York Times today there is an article on how students at Princeton are having trouble getting jobs on Wall Street.  While the overall tenor of the article is depressing, the phrases that include operations research are very nice:

Despite being in the rigorous Operations Research and Financial Engineering program, …

…vast armies of Wall Street recruiters have traditionally taken over the historic Nassau Inn nearby to woo not just Operations Research and Financial Engineering whizzes but innocent art history majors as well.

Sure, it would be better to have been part of an article on how everyone in operations research is getting a $250,000 job and guaranteed life-fulfillment, but having the New York Times readers associate “operations research” with rigor and woo-ability is an association that we literally cannot buy.

So if you are part of a department (either in academia or business) considering replacing “operations research” with a trendier name (“analytics”, “business intelligence”, “information engineering” and so on), you might want to fight that move:  consider the fate of all the groups who went with “eBusiness” and “web” ten years ago.  And those of you who changed to something more “with-it”:  come on back!  There is still room under the umbrella for you.  You can be part of the “operations research” success.

Note added 7PM: I see the tagline on the Edelman Page is “The Best of Applied Analytics”.  Even INFORMS (INFAA?) isn’t immune to trendy names.

What stops you from using Open Source in Operations Research?

In the discussion of CPLEX versus Gurobi software, the discussion took an interesting turn when it came to open source (like that of COIN-OR).  Sebastian, bemoaning the cost of commercial packages such as CPLEX said:

I think another long-term way to make OR feasible in situations where the cost of Cplex cannot be justified is to invest in the long term development of open source alternatives. The COIN-OR solvers are quite good already (for LP and general non-linear programs they are even competitive with state-of-the-art). Moreover the CPL license IBM chose is very liberal and allows the solvers to be tightly integrated into commercial products without further obligations.

If more companies would support the development as a long-term investment it could be in their best interest.

Shiva, an OR practitioner for a decade replied:

Sebastian: We did investigate open-source methods (including COIN), but after reading the fine print, legal departments tend to shoot it down with prejudice to avoid exposing the company to potential lawsuits down the road. It was another frustrating experience, since like all O.R folks, I love what COIN’s been doing.

Larry, of IEOR Tools, replied:

Shiva, Great points on barrier to entry. I too have been caught in that world of justifying an IT expense for the sake of desicion modeling. For instance, one kind find a decision analysis by just using a spreadsheet. Is it optimal? No. Does it offer a solution? Yes. The key is to justify the long term gains of optimal decision processes. I believe that is the trick.

Yet I have definitely found a great niche and I think the OR community can great help with this issue. Open Source software has the added benefit of community support, development, and implementation without the cost structures. It’s something I have started talking about on my blog. I believe it can have profound affects on the OR community.

I repeat all this because this is one of the best exchanges we have had on MTORB.  I am part of COIN-OR (a member of the “Strategic Leadership Board”, though I feel at sea at this compared with my more experienced colleagues).  And it does interest me why more people don’t use COIN-OR.  For me, there are issues with documentation:  I don’t have a lot of time when I learn a new tool, and much open source stuff doesn’t have documentation and examples at the level of commercial software.  I spent a summer working with Symphony, now at COIN-OR, and it was great fun.  I even put together a document on my experience that was part of the distribution at one point.  But I am not sure how many summers I can spend on working through new software.  So I use COIN-OR software but not as much as I should.

I am also interested in why people do use COIN-OR and other open source tools.  I very much welcome thoughts on this issue, and thank Sebastian, Shiva, and Larry for getting this going.

Check out ILOG’s DIALOG blog!

I am in Minneapolis, flying back to Pittsburgh after spending a couple days in Winnipeg for a memorial service for my Mom.  My Mom’s passing has a number of effects, large and small.  On the large side, my son Alexander has lost both his gradmothers in the past six months, which makes me sad:  every kid needs plenty of grandparents to spoil them!  Now my Dad will have to pick up all the slack.

On the small side, I had to cancel a trip I was looking forwad to:  ILOG’s DIALOG conference in Orlando.  First, it is Orlando, which looks pretty sweet for a Pittsburgher in February.  Second, with all of the changes for ILOG in the past year, I was looking forward to meeting people and seeing how the optimization side of ILOG was making out.  No chance to do that, unfortunately, but ILOG has significant blog coverage of the conference, which I recommend.  Most of the guest bloggers are rules-oriented (my absence messes up the optimization covererage, I guess) but it was good to read that the plenary of Tom Rosamilia (head of WebSphere) sees how optimization fits in (from James Taylor’s blog entry):

Tom identified four essentials for survival:

  • Adapt to embrace change
  • Streamline processes to make them more dynamic and manageable
  • Optimize to allocate resources efficiently
  • Visualize to transform insight into action for faster decisions

This is not quite the way I see optimization (either it is much more than resource allocation, or resource allocation has a much broader definition than I usually give it!), but at least we hit the main points.

Constraint Programming Makes the Big Time

The current PhD (Piled Higher and Deeper) comic (a comic written around the lives of doctoral and postdoc students:  I fear I see too much of Dr. Smith in me!) revolves around Constraint Programming as applied to wedding planning.  Karin Petrie, a well known constraint programmer, had lunch with the author of PhD, and I expect this is not a coincidence.

American Express and Data Mining

I teach a data mining course here to our MBA students.  It is a popular course with about 70% of the students taking it at some point during their two years with us.  Since I am an operations research guy, I concentrate on the algorithms, but we spend a lot of time talk on the use of data mining, and possible pitfalls.  The New York Times today has a wonderful story illustrating the pitfalls.  American Express has been using shopping patterns to reduce customer’s credit limits.  This, in itself, is not surprising, but a letter the company sent out implies that it was basing the evaluation on the stores and companies the customer used, rather than a more direct measure of consumer ability to repay a debt.

“Other customers who have used their card at establishments where you recently shopped,” one of those letters said, “have a poor repayment history with American Express.”

Wow!  Shop at the Dollar Mart, you are not a careful shopper reacting to an uncertain financial world, but rather a poor credit risk who should be jettisoned before defaulting (I don’t know if Dollar Mart is one of the “bad” establishments: American Express has not released a list of companies that are signs of imminent financial doom). That is, of course, what data mining results come down to, but it is rare for a company to admit it.  Not surprisingly, customer’s who received such a letter became a little irate.  Check out newcreditrules.com for one person’s story.

American Express says it is no longer using store shopping information, but it will continue to use the results of data mining in its credit decisions.

In one presentation to analysts, it noted that people with multiple residences and multiple mortgages used to be a good bet. Now, the reverse is true.

In a good economy, lots of data mining was used to “help” customers by identifying new products or offers that might appeal to them.  Now, it seems that more data mining uses the customer’s data against their own interests.  I suspect we will see more stories of this type.

Gurobi versus CPLEX benchmarks

Hans Mittelmann has released some benchmark results comparing CPLEX 11.2 with the first version of Gurobi‘s code (1.02) in both sequential and parallel (4 processor) mode.  Mosek’s sequential code is also included in the test.  Let me highlight some of the lines:

=================================================================
s problem     CPLEX1    GUROBI1     MOSEK1     CPLEX4    GUROBI4
-----------------------------------------------------------------
    air04          9         18         49          7         13
   mzzv11         89        116        434         80        116
      lrn         42        104          f         36        996
ns1671066       1828          1       2474         26          1
ns1688347       1531        315          f        716        258
ns1692855       1674        322          f        688        234
     acc5         25        247       4621         30         33

These are not chosen to be typical, so be sure to look at the full list of benchmark results.  Concentrating on the first two columns, Gurobi’s code seems to be holding its own against CPLEX, and sometimes has a surprising result.  The ns1671066 result certainly looks like Gurobi is doing something interesting relative to CPLEX.

It is also striking how often the speedup is more than a factor of 4.  That same ns1671066 instances that CPLEX had such trouble on sequentially certainly looks much more solvable with 4 threads.  But sometimes there is a significant slowdown (see Gurobi on lrn).  Perhaps someone more skilled in this area can explain how we should interpret these benchmarks.

The official release for Gurobi is still two months away:  it will be interesting to see how the released version works.

Over- versus Under-Planning

My wife and I have been arguing recently about my family’s tendancy to over-plan.  The Trick way does tend to have plans with lots of contingencies, which perhaps a reason operations research appeals to me.  The Weyers (my wife Ilona’s surname) approach is a little more … take it as things come.

The Think-OR blog has a wonderful OR joke that pretty well summarizes how us over-planners work.  I think it is instructive that both Ilona and I think that the story vindicates our chosen position.