Diagramming with Dia tutorial

BeginningLinux.com has a great tutorial on the diagramming tool Dia. Dia is a free and open source diagramming tool that draws its inspiration from Microsoft Visio. From the Dia website...

It currently has special objects to help draw entity relationship diagrams, UML diagrams, flowcharts, network diagrams, and many other diagrams. It is also possible to add support for new shapes by writing simple XML files, using a subset of SVG to draw the shape.

It can load and save diagrams to a custom XML format (gzipped by default, to save space), can export diagrams to a number of formats, including EPS, SVG, XFIG, WMF and PNG, and can print diagrams (including ones that span multiple pages).

Dia is available to download for Windows and Linux. The tutorial even explains how to download and install for each operating system.

The Industrial Engineer can really use Dia for a number of projects. Just to name a few include flowcharting process flows, diagramming models of queues, and mind mapping exercises.

The BeginningLinux.com tutorial does a great job of detailing the steps to get projects started. I find that tutorials with pictures or screenshots tend to be the best. This tutorial has plenty of screenshots to guide the beginner diagrammer.

As I have mentioned before there are great alternative IEOR Tools available in the free and open source community. Give Dia a try and let me know about your experience.

The flaw of averages

An interesting article from the Washington Times about a new book "The Flaw of Averages" by Sam L. Savage. Savage says that basing decisions on averages alone can be a dangerous assumption. In fact, the assumption the decision is going to be based is going to be wrong half the time. From the article and I believe this to be well said

It is the flaw of averages that causes businessmen, engineers, generals and others to underestimate risk in the face of uncertainty.

How true is that statement. As a practitioner of Operations Research in the business world I often am not just providing solutions for my employer but also playing referee. I'm not sure how many times I have had to tell a manager or associate to recheck their assumptions before they bring their analysis up the chain of command. One of my previous bosses had a great phrase that he would tell me, "Does it pass the smell test?" In other words can you step back and say with certainty that your assumptions are right on the mark.

Sam Savage has a great point about flawed analysis based on assumptions alone. The crucial emphasis should be on the assumptions themselves. All good analysis should make sure that the assumptions are correct and reflect the real world implications. The article cites the recent problems with the home mortgage crisis. A lot of assumptions in the risk models that gave AAA ratings to mortgage-backed securities were not taking into account future market cycles and risk.

So should averages be used in analysis? By all means, no pun intended. Statistical inferences and generalities require average value of numerical populations. Averages do have their place in analysis when comparing samples. Yet so do standard deviations, variance, correlation, and so on. There is more to risk modeling than just the simple average.

Beware the Zombie attack according to latest models

Thanks to the expert mathematical minds at the University of Ottawa they have found that in the event of a zombie attack we would be wiped out rather quickly. Their answer is to attack "quickly and aggressively" according to the study. Their report was published in the book "Infectious Diseases Modeling Research Progress". Treehugger and Wired also covered this story.

Apparently the modeling research was developed to study unfamiliar diseases. The modeling can be used to find the affects of swine flu and SARS. And in this case apparently from the affects of zombie attack outbreaks! Perhaps we should all be prepared.

Social Networks and optimal contacts

The Open Road blog has an interesting article about interconnecting "meta-address books". In this article Matt Asay explains how open source companies are trying to optimally connect people together by the meta-data they place in the internet cloud such as using Open-Xchange.

One such example they describe is PeopleMaps from 7-Degrees. By leveraging what they know about someone online they can use that data to connect with other people of related interests, business, and other potential connections. According to the 7-Degrees website they use their own industry-leading path-finding algorithms to enable the connections between the users.

Perhaps Operations Research has a big future with social networks.

Army geeks apply Operations Research for intel

Strategypage.com has an article about the Army troops using Operations Research help predict enemy activity. The article is praising the efforts of the Army "geeks" that uses Operations Research tools to find enemy IED's, predict enemy hotbeds, and help plan supply routes and enemy engagements in Afghanistan and Iraq.

The Army uses several different tools to help find particular operations objectives. Pattern analysis and data mining is used, with the aid of digital image intelligence, to spot IED's and those responsible for planting the bombs. Predictive analysis is used to determine supply routes that may have heavy enemy activity. The predictive reports are reported to convoys and explosive disposal units to help quickly deliver operations' supplies.

IBM buys SPSS analytics company

IBM is going further into the business analytics field with the acquisition of SPSS for $1.2 billion as reported by the Wall Street Journal. From the article, IBM is expanding its business analytics focus as the demand increases from its customers for decision-based analysis. On the forefront of client's needs is to cut costs.

This is really interesting and should really boost recognition to field of Operations Research. IBM seems to be positioning itself as the world leader in business analytics. The article says that business analytics is going to grow by 4% to $25 billion world-wide market for the year.

Netflix prize has come to a dramatic end

The Netflix Grand Prize challenge has come to an end and it was a real race to the finish. It looked like the award was going to go to the team of BellKor's Pragmatic Chaos. Yet in a surprise, and perhaps very cunning move, a new team emerged known as The Ensemble.

There is already a lot of buzz about the finish of the competition. Perhaps the first to talk of the finish was in the Operations Research blog-sphere with Sebastian Pokutta's blog. Sebastian did a great job of alerting the final days activities and keeping us up to date.

Among other places on the web include Slashdot's netflix update story. Techcrunch also submitted an article about the final push to the end of the netflix grand prize.

All in all I think this is great exposure to mathematics decision analysis, operations research, and machine learning. Regardless of your thoughts of the outcome I would have to give Netflix a lot of credit for hosting this competition and promoting it publically.