Breakthrough Bullets

Crystal Ball vs ModelRisk in Discrete Distribution Fitting and Correlation/Copulas (8/8)

kluce@technologypartnerz.com — Tue, 22 Feb 2011 16:02:00 GMT

Is there a winner in this battle between Crystal Ball and ModelRisk? To quote that way-too-often-quoted reply: It depends. Some users will value certain technical capabilities over others. Some users will value user-friendliness over accuracy. If there is to be a group deployment of a MCA spreadsheet package, usability may trump technical capabilities overall. Does it matter if one package has more distributions to choose from if there are only three that are of interest for your particular class of stochastic problems? Would it matter what kind of correlation enforcement method is used if, as in many manufactured assemblies, there is practically no correlation between separate components? Probably not. But if they do (as in financial and insurance applications), there will be a clear winner.

Tags: Crystal Ball,ModelRisk,Distribution Fitting

Correlation of Duke Basketball Scores, in ModelRisk (7/8)

kluce@technologypartnerz.com — Mon, 14 Feb 2011 15:56:00 GMT

Correlation behavior in ModelRisk is enforced with the use of copulas. Copulas offer more flexibility in accurately simulating real data scatter-plot patterns than do single-value correlation coefficients. While this advantage is clear for financial and insurance applications, its implementation in an MCA spreadsheet simulator can make the difference between universal adoption and rejection by a majority of the intended user group. Let us now use ModelRisk (MR) to enforce the correlation behavior between Duke Basketball offense scores and their opponents' scores, based on the '09/'10 historical data.

Tags: Crystal Ball,ModelRisk,Distribution Fitting

Correlation of Duke Basketball Scores, in Crystal Ball (6/8)

kluce@technologypartnerz.com — Thu, 03 Feb 2011 16:59:00 GMT

In our quest to simulate future Duke Basketball scores, we have taken past historical data of individual games during the '09/'10 season and fitted probability distributions to that data. Two PDFs are generated; one for Duke's scores (offense) and one for their opponents' scores (defense). We have used both Crystal Ball and ModelRisk to perform this task. Is there something missing in our PDF formulations?

Tags: Crystal Ball,ModelRisk,Correlation

Correlation and Impact on Monte Carlo Analysis Results (5/8)

kluce@technologypartnerz.com — Thu, 27 Jan 2011 17:54:00 GMT

All the top dogs in the Monte Carlo Analysis spreadsheet universe have distribution-fitting capabilities. Their interfaces have common elements, of course, since they rely on (for the most part) the same PDFs in their arsenal of distribution-fitters. There are important differences, to be sure. It is hoped this comparison will illustrate pros and cons from a practical standpoint. Before going over our scorecard between Crystal Ball and ModelRisk, there is one more very important capability category begging for review: Correlation.

Tags: Correlation,Monte-Carlo,Statistics

Discrete Distribution Fitting to Duke Basketball Scores, in ModelRisk (4/8)

kluce@technologypartnerz.com — Thu, 20 Jan 2011 17:10:00 GMT

Let the battle begin anew. We continue our journey in uncertainty modeling, having understood how to fit distributions to data using Crystal Ball (CB). How does that experience compare to what ModelRisk (MR) has to offer?

Open the Duke 09_10 Scores spreadsheet with ModelRisk loaded in the Excel environment. We will first create the MR Objects representing the fitted PDFs. (Just as with the CB exercise, it is good practice to examine a variety of best-fitting distributions, rather than blindly accepting the top dog.) Then, in distinctly separate cells, we will create the VoseSimulate functions that behave as sampled values from the PDFs modeled by the MR Objects.

Tags: Distribution Fitting,ModelRisk,Statistics

Distributions in ModelRisk as Objects (3/8)

kluce@technologypartnerz.com — Wed, 12 Jan 2011 15:02:00 GMT

As with Crystal Ball, ModelRisk has the ability to fit distributions to historical data. The analyst looking to describe the variation of a Monte Carlo Analysis input can use "fitting" windows to select data and manipulate other options. How does the ModelRisk (MR) fitting experience stack up against the Crystal Ball (CB) methods and options? There are some important differences one should understand about MR before fitting PDFs to the Duke 09_10 Scores spreadsheet.

Tags: ModelRisk,Crystal Ball,Statistics

Subject Matter Expertise in Distribution Selection (2/8)

kluce@technologypartnerz.com — Thu, 23 Dec 2010 17:05:00 GMT

Are there discrete univariate probability distribution functions (PDFs) that can be used to simulate college basketball scores? Do we, as avid basketball observers, know enough to suggest one discrete PDF is better than another? In fitting distributions to data in your business problems, the analyst will be asking the same types of questions. If the analyst is not an expert on the inputs and their behavior, he or she should seek out a subject-matter expert (SME) who can provide insight. Putting experience and theoretical knowledge together this way is a best practice for distribution selection.

Tags: Statistics,Distribution Fitting

Discrete Distribution Fitting to Duke Basketball Scores, in Crystal Ball (1/8)

kluce@technologypartnerz.com — Thu, 16 Dec 2010 16:19:00 GMT

Let us assume we have a batch of historical data in a spreadsheet. Our mission-of-the-moment is to use this data and fit probability distributions that describe its past variability (or uncertainty). Consider using either Crystal Ball or ModelRisk to do this task. We offer free trials of both to registered users. If you register here, you can get yours too. Try fitting the same data using these two different packages. Let us know how and why one is better than the other. In demonstrating these capabilities, we gain first-hand experience on the usability and capabilities of the alternatives and which features compared have more priority. The best way to judge is to try them out for yourself.

Tags: Distribution Fitting,ModelRisk,Statistics,Crystal Ball

Dealing with Uncertainty

kluce@technologypartnerz.com — Thu, 09 Dec 2010 20:08:00 GMT

Change is constant. Or so the saying goes. However, even change is ever-varying. So perhaps we should say: Change is constantly changing. As occupants of planet earth, we intuitively know this and yet strive to keep everything the same, at least those things that do well by us. Uncertainty derails the best of our plans, even uncertainties that we recognize up front.

Tags: Statistics,Monte-Carlo,Risk Analysis

Tolerance Analysis Summary (Part 13 / 13)

kluce@technologypartnerz.com — Thu, 04 Nov 2010 18:01:00 GMT

Tolerance Analysis focuses on dimensional aspects of manufactured physical products and the process of determining appropriate tolerances (read: allowable variations) so that things fit together and work the way they are supposed to. When done properly in conjunction with known manufacturing capabilities, products don't feel sloppy nor inappropriately "tight" (i.e., higher operating efforts) to the customer. The manufacturer also minimizes the no-build scenario and spends less time (and money) in assembly, where workers are trying to force sloppy parts together. Defects are less frequent. There are a wealth of benefits too numerous to list but obvious nonetheless. Let us measure twice and cut once.

Tags: Engineering,6 sigma

Tolerance Analysis using Monte Carlo, continued (Part 12 / 13)

kluce@technologypartnerz.com — Mon, 01 Nov 2010 16:33:00 GMT

In the case of the one-way clutch example, the current MC quality prediction for system outputs provide us with approximately 3- and 6-sigma capabilities (Z-scores). What if a sigma score of three is not good enough? What does the design engineer do to the input standard deviations to comply with a 6 sigma directive?

Tags: Engineering,6 sigma,Monte-Carlo,Statistics,Simulation

Tolerance Analysis using Monte Carlo (Part 11 / 13)

kluce@technologypartnerz.com — Thu, 28 Oct 2010 15:02:00 GMT

How do Monte Carlo analysis results differ from those derived via WCA or RSS methodologies? Let us return to the one-way clutch example and provide a practical comparison in terms of a non-linear response. From the previous posts, we recall that there are two system outputs of interest: stop angle and spring gap. These outputs are described mathematically with response equations, as transfer functions of the inputs.

Tags: Engineering,6 sigma,Monte-Carlo,Statistics,Simulation

Introduction to Monte Carlo Analysis (Part 10 / 13)

kluce@technologypartnerz.com — Mon, 25 Oct 2010 22:26:00 GMT

In past blogs, I have waxed eloquent about two traditional methods of performing Tolerance Analysis, the Worst Case Analysis and the Root Sum Squares. With the advent of ever-more-powerful processors and the increasing importance engineering organizations place on transfer functions, the next logical step is to use these resources and predict system variation with Monte Carlo Analysis.

Tags: Engineering,6 sigma,Monte-Carlo,Statistics,Simulation

Root Sum Squares Explained Graphically, continued (Part 9 / 13)

kluce@technologypartnerz.com — Fri, 24 Sep 2010 14:00:00 GMT

The other RSS equation, that of predicted output mean, has a term dependent on 2^nd derivatives that is initially non-intuitive:

Why is that second term there?

Tags: Engineering,6 sigma

Root Sum Squares Explained Graphically (Part 8 / 13)

kluce@technologypartnerz.com — Tue, 21 Sep 2010 18:42:00 GMT

A few posts ago, I explained the nature of transfer functions and response surfaces and how they impact variational studies when non-linearities are concerned. Now that we have the context of the RSS equations in hand, let us examine the behavior of transfer functions more thoroughly.

Tags: Engineering,6 sigma

Tolerance Analysis using Root Sum Squares Approach, continued (Part 7 / 13)

kluce@technologypartnerz.com — Wed, 01 Sep 2010 22:07:00 GMT

As stated before, the first derivative of the transfer function with respect to a particular input quantifies how sensitive the output is to that input. However, it is important to recognize that Sensitivity does not equal Sensitivity Contribution. To assign a percentage variation contribution from any one input, one must look towards the RSS output variance (σ_Y²) equation:

Tolerance Analysis using Root Sum Squares Approach (Part 6 / 13)

kluce@technologypartnerz.com — Mon, 30 Aug 2010 21:00:00 GMT

Root Sum Squares (RSS) approach to Tolerance Analysis has solid a foundation in capturing the effects of variation. In the days of the golden abacus, there were no super-fast processors willing to calculate the multiple output possibilities in a matter of seconds (as can be done with Monte Carlo simulators on our laptops). It has its merits and faults but is generally a good approach to predicting output variation when the responses are fairly linear and input variation approaches normality. That is the case for plenty of Tolerance Analysis dimensional responses so we will utilize this method on our non-linear case of the one-way clutch.

Tags: Engineering,6 sigma

Transfer Functions & Response Surfaces in Tolerance Analysis (Part 5 / 13)

kluce@technologypartnerz.com — Mon, 23 Aug 2010 19:36:00 GMT

Transfer Functions (or Response Equations) are useful to understand the "wherefores" of your system outputs. The danger with a good many is that they are not accurate. ("All models are wrong, some are useful.") Thankfully, the very nature of Tolerance Analysis variables (dimensions) makes the models considered here concrete and accurate enough. We can tinker with their input values (both nominals and variance) and determine what quality levels may be achieved with our system when judged against spec limits. That is some powerful stuff!

Tags: Engineering,6 sigma

Probability Distributions in Tolerance Analysis (Part 4 / 13)

kluce@technologypartnerz.com — Thu, 19 Aug 2010 20:11:00 GMT

With uncertainty and risk lurking around every corner, it is incumbent on us to account for it in our forward business projections, whether those predictions are financially-based or engineering-centric. For the design engineer, he may be expressing dimensional variance in terms of a tolerance around his nominal dimensions. But what does this mean? Does a simple range between upper and lower values accurately describe the variation?

Tags: Statistics,Distribution Fitting,Engineering

Tolerance Analysis using Worst Case Approach, continued (Part 3 / 13)

kluce@technologypartnerz.com — Mon, 16 Aug 2010 19:50:00 GMT

In my last couple of posts, I provided an introduction into the topic of Tolerance Analysis, relaying its importance in doing upfront homework before making physical products. I demonstrated the WCA method for calculating extreme gap value possibilities. Implicit in the underlying calculations was a transfer function (or mathematical relationship) between the system inputs and the output, between the independent variables and the dependent variable. In order to describe the other two methods of allocating tolerances, it is necessary to define and understand the underlying transfer functions.

Tags: Engineering,6 sigma

Tolerance Analysis using Worst Case Approach (Part 2 / 13)

kluce@technologypartnerz.com — Thu, 12 Aug 2010 20:53:00 GMT

As stated in my last post, there are three common approaches to performing Tolerance Analysis. Let us describe the simplest of the three, the Worst Case Analysis (WCA) approach. An engineering-centric term in the Tolerance Analysis world would be Tolerance Stacks, usually meaning in a one-dimensional sense. The explanation begins with probably the most overworked example found in dusty tomes (my apologies in advance).

(I would like acknowledge James Ministrelli, DFSS Master Black Belt and GD&T Guru Extraordinaire, for his help & advice in these posts. Thanks, Jim!)

Tags: Engineering,6 sigma

Introduction to Tolerance Analysis (Part 1 / 13)

kluce@technologypartnerz.com — Tue, 10 Aug 2010 19:57:00 GMT

Tolerance Analysis is the set of activities, the up-front design planning and coordination between many parties (suppliers & customers), that ensure manufactured physical parts fit together the way they are meant to. Knowing that dimensional variation is the enemy, design engineers need to perform Tolerance Analysis before any drill bit is brought to raw metal, before any pellets are dropped in the hopper to mold the first part. Or, as the old carpenter's adage goes: Measure twice, cut once. 'Cause once all the parts are made, it would be unpleasant to find they don't go together. Not a good thing.

Tags: Engineering,6 sigma

Risk Identification and Mitigation: A Engineer’s Perspective

kluce@technologypartnerz.com — Fri, 18 Jun 2010 14:35:00 GMT

The recent and ongoing disaster in the Gulf of Mexico has raised some questions as to the preparedness of Big Oil to respond to an emergency spill. An examination of their emergency spill plans has garnered criticism but is it justified? How much effort should companies spend on risk identification and mitigation?

Tags: Engineering,Risk Analysis