The developerWorks Connections Platform is now in readonly mode and content is only available for viewing. No new wiki pages, posts, or messages may be added. Please see our FAQ for more information. The developerWorks Connections platform will officially shut down on March 31, 2020 and content will no longer be available. More details available on our FAQ. (Read in Japanese.)
2015 Prediction: Prescriptive Analytics Will Make It
This is prediction season. I never played that game so far, but felt compelled to do so after reading quite a few predictions about what will happen in 2015 around Analytics and Big Data. I won't repeat what seems to be a consensus, and will refer to two specific lists that I found more interesting than others. The first list is by Nathan Brixius, from Microsoft. Here are the top items, I'll let you read Nathan's blog to get the meat behind the titles. Adoption of higher productivity analytics programming... [More]
Tags: big_data analytics optimization prescriptive 
Analytics Is A Mean To An End
Unless you've been unplugged for a couple of years, you have certainly witnessed the buzz around Big Data and Analytics . It is difficult to open a newspaper without seeing references to it on a monthly, if not weekly, basis. This is understandable, given the spread of analytics applications. It is now becoming common knowledge that analytics can help, whatever activity you are engaged with. The more I read, the more I see new analytics projects at IBM or at customer companies, the more I feel compelled to... [More]
Tags: big_data optimization analytics 
CPLEX 12.5.1
Update on June 28 2015. A more recent release of CPLEX is now available, namely CPLEX 12.6.2 . We are about to have a new CPLEX release, namely IBM ILOG CPLEX Optimization Studio 12.5.1. We plan for electronic availability this week. The main focus for this release was to improve performance across the board, and particularly for Mixed Integer Programming (MIP) and Mixed Integer Quadratic Programming (MIQP and MIQCP). We also improved numerical accuracy and fixed few bugs. For MIP we get the following geometric average... [More]

Making Dreams Come True (Or Not)
Last week, in my What Is The Solution When There Is No Solution entry, I commented about the fact that not all business requirements were actual constraints and that some of these are probably wishes that business people dream to fulfill. Making the difference between wishes and hard constraints is key to success. Indeed, handling wishes as hard constraints may result in infeasible models, which would be useless to the business people that come to you for help. It turns out that there is a surprisingly simple way to detect that some... [More]
Tags: optimization modeling analytics 
Knowing The Optimum Helps A Bit
Recent experiments  An updated version is available here  by Rodolfo Carvajal show that knowing the optimal solution does not dramatically change the difficulty of mixed integer programming (MIP) problems. In his experiments, Carvajal found that providing an optimal solution at the start of computation only speeds up CPLEX 12.4 by 2 to 4 times on average. This is reminiscent of another experiment by Tobias Achterberg mentioned here : providing the optimal solution speeds up CPLEX MIP by about 2 on average. A 2x speedup is... [More]

IT Best Kept Secret
What is IT best kept secret? I am pretty sure some unknown jewels come to mind immediately, but I am pretty sure as well that few have documented ROI as important as those provided by Mathematical Optimization (aka Operations Research, OR in short). For instance, check this user case where a utility saves over $2Billion. Isn't it worth spreading the news? Shouldn't every utility jump on a technology that can save billions? Well ... many don't. So what? One could wonder why I'm making such a fuss about the utility industry not using... [More]
Tags: orms analytics optimization o.r. 
Un Peu de Math
The following was triggered by a mathematical problem proposed by Vincent Granville. The problem was to compute the maximum value q(n) of a function related to the well known metrics Spearman's footrule , or L1. This would then be used in a new stat isti cal corr elat io n based on ranked variables that would be very useful for Big Data applications. I'll refer readers to Granville's article for more details. At first sight this seemed quite diffficult, and Granville launched a... [More]
Tags: mathematics big_data analytics 
Interactive Optimization
In my last post I discussed how gamification could be used to overcome the resistance to automated decision making systems. The case discussed in my previous post was about a system that computes retail prices for hotel rooms. The point of gaming was to show that human intervention would degrade the business outcome. Prices set by humans lead to less revenue than prices set by the system. Interesting comments on that post made me realize that I have been a bit extreme in my will to make a point. While I stand by my... [More]
Tags: optimization decision analytics 
2013
To all my readers, I hope you enjoyed reading this blog as much as I enjoyed writing it. I want to thank you all as you are the reason for the existence of the blog. You'll find below a quick recap of the blog activity for 2013. I wish all of you a happy and fruitful 2014! My blog got about 130,000 page views (not counting the home page) during 2013, with 31 new posts. It is a significant increase compared to the 31,000 pages views in 2012. The number of views increased significantly very late in the year with the... [More]
Tags: analytics cplex optimization 
Decision makers need decision support
How can you make an optimization application be accepted by decision makers? The answer I gave that in my last post was to provide interactive applications. It so happens that colleagues of mine already discussed that in an IBM book Optimization and Decision Support Design Guide I can't resist quoting some of it given how it captures what I tried to express in my previous post. Decision makers need decision support Decision makers will not use any analytics tool unless they trust it. Trust arises... [More]
Tags: decision optimization analytics 
Financial Institutions Use CPLEX On Mainframe To Manage Risks And Improve Efficiency
Last year we announced the availability of our CPLEX mathematical solver on IBM mainframes called z Enterprise. Since then we have seen financial institutions use it in various ways. This article by my colleagues Aimee EmeryOrtiz, Ferenc Katai, and Sofiane Oussedik review some of these use cases. The paper provides some insights on how CPLEX can be used on mainframe to address banking and finance IT challenges. Not only some business operations need to be optimized, but this has to be done while maintaining customer data... [More]
Tags: analytics zenterprise 
Do More With Less
Most past use of Operations Research (OR) were about lowering cost, or other resource usage metrics. We now see more and more use of OR to generate more revenue, see for instance this example There is also a middle ground, where OR is used to achieve more while using less resources. In short, OR is used to do more with less . A great example is described in this video, by a very enthusiastic user of analytics and optimization. Facing an aging infrastructure and numerous customer complaints, the District of... [More]
Tags: smarter_water optimization analytics 
ODME 3.7 Announce
We also just announced a new release of our IBM ILOG ODM Enterprise product. This product complements CPLEX by providing a platform for advanced analytics solutions based on
optimization, with outofthebox features such as 'whatif' analysis,
userfriendly GUIs, and a repeatable implementation process for
deploying enterprisewide, decision support. It includes features like a business data model, business user interfaces, client server architecture, data server, support for whatif analysis at the business level, etc. These... [More]

Accountable Analytics
My post on How Does Cognitive Computing Relate To Analytics? triggered some interesting discussion. It all started with this tweet. The discussion was interesting, and it quickly moved to the issue of responsibility when decision making is fully automated. If prescriptive analytics computes decisions that are automatically executed, then what is the recourse when something bad happens? Think of self driving cars like Google cars. Who is responsible if such car causes an accident? One... [More]

When Better Is Not Better
I don't know about you, but my experience is that all customers want the best solution to their business problem. And they want it rapidly, within some given time limit. I can't blame them, I do the same when I am the customer. However, if we have finite computing resources then there is a limit to what can be computed. If the business problem is so complex that it can't be solved exactly in the given time limit, then optimization won't compute the best possible solution (one among the optimal ones). What it will compute are good feasible... [More]
Tags: analytics optimization cplex 
Discovering IBM
I just joined IBM as an employee of the former ILOG company that IBM just acquired. Within ILOG I was in charge of the development of ILOG mathematical optimization products (CPLEX, CP Optimizer, OPL, and ODM). I will continue in this role within the AIM division of IBM Software Group. Joining IBM is very exciting because IBM has a long story in the mathematical optimization field. Currently IBM has a strong group in IBM Research. It also has just launched a new initiative in its Global Business Service division, namely BAO (Business... [More]
Tags: optimization ilog 
Interesting Optimization Blogs
Mathematical optimization is often called Operations Research (OR). I'll keep using mathematical optimization, or optimization in short, here, but if you want to learn more about it look for OR resources, for instance OR blogs. The good news here is that Mike Trick is maintaining a list of interesting blogs on his home page: http://mat.tepper.cmu.edu/blog/ Look for the right column, and you'll find a very interesting list. I won't reproduce it here, but all are worth a look at. It is truly a great list. Note that I am not saying this... [More]
Tags: blogs optimization learning 
Do We Need Accuracy In Solvers?
I just read this interesting blog post by Samik Raychaudhuri on solver accuracy. He wrote : " typical data accuracy does not go beyond the decimal point, and even if it does, seldom beyond 12 places of decimal. Why solve a problem accurately up to 10 places of decimal when the input data is not that accurate? " It reminded me of the 53 bits ought to be enough for anybody post by John D. Cook. Paul Rubin commented on the first post that one of the problems that leads to accuracy... [More]

Computing The Longest Tour Across The USA
My previous post Computing The Really Optimal Tour Across The USA On The Cloud With Python generated some interest (like all answer's to Randy Olson's original article ). One reader asked a seemingly simple question that got my interest: My first reaction was to ask what was the definition of a maximal tour? Indeed, without further information, we can get tour of arbitrary length by circling. For instance, let us consider two point only. What would be the longest route between those two points? Well,... [More]
Tags: python optimization analytics 
Start With History?
Should we start with building history data before we know what we collect data for? Or should we really start with a question before collecting any data? I recently advocated for the latter in Start With A Question . I am not the first one, by far. For instance W. Edwards Deming wrote about it in 1942. While many agreed with this message, and even rephrased it in better ways than mine, Paul Rubin pointed to a kind of contradiction. In his words: I do not entirely disagree with the Deming and Godin quotes,... [More]

Decision Optimization In The Cloud at INFORMS Annual Meeting
Update on Jan 10, 2015. Decision Optimization on Cloud open beta is now available. I documented step by step on boarding here, but you can also try by yourself here . Trying Decision Optimization on Cloud beta: part 1, Demo Trying Decision Optimization on Cloud beta: part 2, CPLEX Trying Decision Optimization on Cloud beta: part 3, OPL Our INFORMS conference slides are available at our INFORMS 2014 Annual Meeting page Original post: The IBM Decision Optimization team will be quite active at the forthcoming... [More]
Tags: analytics cplex optimization cloud 
Lessons Learned
I had the honor to be invited to give a talk by the Or Complete group on a topic of my choice during a Google Plus hangout . I decided to speak about lessons learned when selling optimization to business users. Readers of this blog won't be surprised, as this is what this blog is about. Slides are available on slideshare Slide show plus voice recording is available on youtube thanks to OR complete team. Note that I had a flu, hence my voice is not my usual one, and the flow is very slow... I covered some topics already covered in this... [More]
Tags: change optimization business 
I Want The Best Solution
Who among optimization practitioners hasn't heard at least one customer saying "I want the best solution" ? I certainly did. When that happens one has to manage expectations because solving real problems to optimality may not be possible in a reasonable time. Customers must understand that finding the best solution and prove it is the best one is not always doable. They should instead focus on finding the best possible solution in a limited amount of time. If they agree to it, then they find it valuable to... [More]
Tags: optimization moov analytics 
Looking Under The Hood
Go to market (GTM) is something you must understand very quickly if you try to live from selling what you do. For optimization, we converged on a very simple view at IBM. There are two categories of customers, those who understand the technology, and those who don't. The formers are operations research (OR) practitioners and OR academics. The latter are line of business (LOB) people. Understanding how the technology work is not so much important to them. What is important is to understand what it enables. The OR market is well served by... [More]
Tags: optimization modeling 
Efficiency Can Get You In Trouble
This post was triggered by a recent tweet by Laura McLay. Laura was pointing to John D. Cook entry on how Efficiency could land you in jail . In that post John explains how a German postman was almost fired because he found better (read more efficient) ways to deliver mail. The story may be a bit more complex (read Chris Nahr's comment on John's post), but it seems plausible. Basically, the postman was punished because he was breaking some glass in order to gain efficiency. In short,... [More]

2012 Edelman Award Recipient Uses CPLEX
I didn't had the chance to attend the 2012 INFORMS Conference on Business Analytics & Operations Research where TNT Express, Ortec and Tilburg University won the prestigious 2012 Edelman Award. I've discussed Edelman award extensively here . One OR expert on the winning team told my colleague Jeremy Bloom that they were using CPLEX along with other unnamed tools. We will deep dive on the use of CPLEX here when details will have been published. It is worth noting that beside adopting OR tools, TNT Express has teamed up with the... [More]
Tags: optimization analytics edelman cplex 
The Executive Guide To Optimization
Steve Sashihara, CEO of Princeton Consulting, has authored a book that readers of this blog should like: The Optimization Edge For those who haven't read it yet (and those who have read it for that matter) I suggest you attend a forthcoming Information week webcast on Thursday, February 21, 2013 at 11:00 AM PT / 2:00 PM ET . Here is the announcement text: The Executive Guide to Optimization Watch
this webcast and learn how to efficiently reallocate your resources and
make timesensitive, complex decisions that drive up market share and... [More]
Tags: optimization book 
Common Sense Operations
Is Optimization the tool of choice for improving operations? I tend to say yes, because we can be proud of the impact of optimization on various businesses and industries. See for instance the Franz Edelman Award finalists for examples of tremendous achievements . Yet, I regularly see nice alternative to using optimization techniques being successfully used. Let me give two examples. I have been visiting a trucking company a while ago, pitching optimization as a way to significantly decrease mileage and fuel... [More]
Tags: analytics optimization 
Un Peu de Math avec CPLEX
There is one thing I didn't disclose in my previous post (Un Peu de Math) on the Analytic Bridge Mathematical Competition : I used CPLEX to refine and validate various hypothesis. As usual, my first try wasn't very convincing, and I had to rely on better understanding of the problem to solve it efficiently. Let me start with a statement of the optimization problem to be solved. A permutation x is a function such that x(i) is in the set {0,1,...,n1} for all i in that set.... [More]
Tags: modeling mathematics 
More On Absolute Value
I needed to model an absolute value in a MIP model. I could have used one of the methods described in the very good review by Paul Rubin , but another method seemed better suited to my special case. I am given a binary variable z and an integer variable x such that 0 <= x <= 2 z = abs(x1) The rest of the model is such that we cannot assume we minimize z. This is therefore a non convex constraint. The method I used works for any function of x . In order to model z = f(x) we introduce one binary x i... [More]
Tags: modeling 