Imperfect Information: The Decision Analysis Blog

Syncopation Software is pleased to announce the general availability of DPL Release 9. This release serves to underscore Syncopation's commitment to providing users with best-in-class decision analytic tools that improve performance, enhance interoperability, and give you insights that lead to better decision quality. Whether you choose the Professional, Enterprise or Portfolio version, we're confident you'll find DPL 9 the best decision analysis software tool you've ever used.

Decision analysis is most often used on high stakes, one-of-a-kind decisions. However, the same techniques and tools can be used to shed light on a variety of decisions, including those where the stakes are, say, "A BRAND NEW CAR!!!".

If you are subscribed to our mailing list this quote may sound familiar to you:

"Enhancements in this [DPL 9] release range from game-changing analytics to modeling ease-of-use to support for the latest technology platforms."

The vast majority of DPL users are also heavy Excel users, and they sometimes ask us about the pros and cons of the various Excel versions. Oftentimes they have no choice in the matter -- the corporate IT department is rolling out a new version of Office, and they just need to know what they're in for. DPL has been closely integrated with Excel from the earliest days, so we definitely have an opinion. This post summarizes the highs and lows of recent Excel history, from the perspective of power users doing analytical work.

I'm going to use a light-hearted example to introduce a simple but powerful new feature in DPL 9: Perform and Continue. If you've used DPL, you probably know about the perform feature, a way of creating "perform links" which reduce the redundancy in a decision tree. Often when building an asymmetric tree there are sections which are repeated -- some groups of nodes are the same on different paths.

Consider the personal decision problem of what to drink. Beer and wine are both enjoyable, but they involve certain risks.

A customer once inquired about a new feature request for DPL 9. It went a little something like this:

You know how DPL does that really cool Initial decision alternatives chart? The one where you can generate risk profiles -- not just for the optimal decision alternative -- but for all of the initial decision alternatives? And, they're displayed conveniently in one chart for easy comparison? Like this:

You know the traditional holiday tune where one delivers increasingly grander gifts to his/her companion over the span of 12 days. I believe it draws a parallel to how one will feel upon starting up our forthcoming release DPL 9 -- licenses for which are now available for pre-release purchase at a 25% discount! It's the software tool that keeps giving when it comes to decision analytics, performance, interoperability and polish. So I give you Syncopation's interpretation of this holiday classic.

A common and challenging decision problem in the oil & gas industry is to decide how to explore an oil field. Typically there is a cluster of prospects, giving rise to a highly dependent set of uncertainties, according to their proximity and the geology of the area. For example, if you drill a well at one site and strike oil, it's more likely that a nearby prospect will also have oil. If you drill a few "dry holes", you'll probably give up on the field rather than throw good money after bad drilling the other prospects. This makes intuitive sense, but until recently it was tough to model in a decision tree.

"…the problem with resource allocation isn’t ignorance of its importance—83 percent of executives we polled named it as the most critical management lever for spurring growth." - McKinsey (Where, how much, and how: Answering the hardest questions of resource allocation)

And yet, McKinsey and Company go on to point out that a third of organizations included in a survey reallocate just 1% of capital from year to year. Whether you're talking about money, talent, or time these precious, finite resources tend to be handled by management teams with far too much safeguarding and rigidity -- leaving them in the dust of those who pursue a more deliberately dynamic resource allocation strategy.

As a market leader in decision analytic tools, we confidently stand by our claim that DPL is the most powerful, full featured application for decision tree and influence diagram modeling on the market today. DPL offers a dedicated, standalone graphical modeling interface for performing decision and risk analyses, Monte Carlo simulation, and Real Option valuation. The tool's flexibility and intuitiveness allow you to model the characteristics specific to your decisions with a precision unmatched by our competitors.

Syncopation's recent Fed Rate Hike promotion not only offers you an opportunity to save money on new or upgraded licenses of DPL, it also gives us a platform to discuss important decision making principles, like the value of information.

One of the most powerful facets of a Decision Analysis approach is the ability to explicitly calculate the value of information so one can make an intelligent choice about whether or not to buy it. If you've read our blog before you should already know the following wisdom about VOI:

After much ado the Federal Reserve decided to maintain interest rates at their current levels this past Wednesday -- which wasn't much of a surprise to most. The Syncopation team saw an opportunity to drum up the drama surrounding the Fed's upcoming November meeting by offering potentially huge savings on DPL licenses, with the amount of savings depending on the outcome of the FOMC's Nov 2 meeting. As a bonus, we're also providing a quick lesson on the value of imperfect information through our promotion. Read on for all the details.

The DPL maintenance update 8.00.14 is now available for download. It includes the following:

BUG: Invalid node data after changing a conditioned value node to a discrete chance node (#2010)
BUG: Spurious "too few initializers" error running a model with init links (#2001)
BUG: DPL may stop responding compiling a model with hundreds of DDE linked nodes (#1997)
CHG: Default stack size increased to 16M (necessary for some very complex model structures)

Ears always tingle in the decision analytics community when a mainstream publication talks about the things we do, and last Friday, the stimulus was provided by this article in the Wall Street Journal on Monte Carlo. Usually I reluctantly read these pieces fully expecting to cringe at the reinforcement of misconceptions and/or zombie fallacies, but in this instance I have to admit I was positively surprised.

In an earlier post I argued that the #1 cause of downward project revisions in portfolio analysis is a phenomenon called "winner's curse", whereby the error in estimating project values, together with screening criteria, results in more negative surprises than positive ones. In this post, I'll talk about the #2 reason (which might be the #1 reason in some portfolios). This reason is more straightforward, it's about what we mean by success.

Deploying a fully integrated PPM System within your organization doesn't have to be expensive, time-consuming, or painful! Embed Syncopation Software's DPMX System into your resource planning processes and start reaping the benefits a straightforward, cost-efficient portfolio analytic system backed by rigorous decision analytics and robust data management tools can bring to your team and portfolio.

The management science news of the week is that Howard Raiffa has died.

Raiffa was one of the founders of the field of Decision Analysis, and indeed of Game Theory and Bayesian Statistics.

In a previous post, Chris expounded on the methods for generating good tornado diagrams. It's relatively straightforward to create a meaningful tornado for a business problem defined by a few symmetric quantitative chance events (i.e., the kind with 10/50/90 assessments and Low/Nominal/High outcomes).

In this space we often write about the do's and don'ts of probability assessment and working with uncertainty in general. One of the most fundamental don'ts is that you should not use words to define probabilities. You should more than avoid it, you should never do it. A 15% chance of something happening is exactly that: a 15% chance. Write it as 0.15 if you like, but don't call it a "moderate risk" or say it "probably won't happen".

Are you a die-hard decision tree fan? Well then you've landed on the right blog. Let me first say that we here at Syncopation (publishers of class-leading decision tree software, DPL) have always prided ourselves on the fact that DPL's modeling interface offers a unique synergism of a Decision Tree and an Influence Diagram. We still hold those sentiments, but during the development of the latest DPL release we came to the realization that for certain users and/or decision-problems the Influence Diagram might be better off taking a backseat.