Review: Full Monte Schedule Risk Analysis for Microsoft Project

02/07/12
by Ten Six

Full Monte is an Add-in for Microsoft Project (MSP), which means that once installed (a breeze, by the way) it appears as part of the MSP menu system.  It works with MSP 2007 and 2010 adding the functionality of Cost and Schedule Risk Analysis using Monte Carlo simulation.  It also comes with a comprehensive User Guide and a shorter “Guided Tour” designed to get you started quickly using a demonstration project supplied with the install.

For those wanting to better understand what cost and schedule risk analysis is all about, then you may want to read a previous blog here called Schedule Risk Analysis: What Is It and Why Do It?

Risky Business

The purpose of risk analysis is to get a more realistic picture of likely project outcomes by recognizing that duration estimates are not precise.  It does this by replacing the single point (or “deterministic”) estimate of the duration of each task with a probability distribution.  The Monte Carlo simulation technique consists of sampling from these distributions and doing the CPM calculations multiple times, or trials, and summarizing the results in terms of the distributions of all the dates, slacks, etc.

The user selects the type and parameters of these distributions using Full Monte’s edit dialog:

Full Monte supports Normal, Lognormal, Beta, Triangular, and Uniform distributions.  Optimistic and pessimistic duration estimates can be entered either as durations or as percentages of the deterministic duration.  Tasks whose duration is influenced by outside events like rain can be correlated by relating them to this outside event.  Full Monte also supports branching, so that only one of a set of alternative paths may actually be operative during a given trial.

There is also a global edit facility, which allows you to set the parameters for all tasks, or for a selected group of tasks.

Once the data is entered, performing the simulation is easy.  There are a few options on the dialog, which sets it up, but most can be left at their default values:

I did 10,000 trials, which took about 15 seconds on the 100-task demonstration project.  This is fast, much faster than I have seen on Risk+ for example.  At the end, it automatically displayed a histogram of the project finish date:

This was one of the options on the dialog, but note that even if you do not opt to see this automatically when the simulation is complete, it still generates the data and the report can be viewed at any time.  It’s also worth noting that the report can be customized, printed, or copied to the clipboard for insertion in other documents.

Reporting

That’s pretty much all there is to it, except for reporting.  All of the results are stored in MSP’s user fields so in theory they can be viewed in MSP, but some of them are encoded in complicated ways and there is also the issue of which user field represents which data.  More on this later.

Full Monte supplies its own reporting function and although there are not many standard reports, the list can easily be expanded by creating your own, which is easy to do.  I particularly liked the ability to add a column for, say, the 95th percentile of the finish date of each task (This is the date by which one can be 95% sure the task will be finished).  Many regular MSP fields can also be included in a Full Monte report.  You can also tailor the bar charts.  One feature I especially liked is when you double click on any row, it brings up a dialog with all the Full Monte results for that task:

Clicking on the “Graph” buttons brings up the corresponding graph — like the one shown earlier but for this individual task — instantly.   There is also a button to show the MSP field names, which can be useful if you want to view the data in MSP, and this brings me to my penultimate point.

The use of user fields could be dangerous if you have data from other applications using those same fields.  Full Monte not only gives you the opportunity to specify which fields to use, so as to avoid conflicts, it actually warns you when there are conflicts.  The first time you use a particular project with Full Monte it looks in all the fields it is expecting to use and warns you if they already contain data.  I tested this by creating a new project and putting some data in Text1 before entering Full Monte.  This was the result:

Finally, there is a utility to transfer data from Risk+.  I was not able to try this for this review.

Summary

Full Monte is an easy to use, fast and robust Add-in for Microsoft Project.  It provides a simple way to add powerful Cost and Schedule Risk Analysis capabilities directly to Microsoft Project. Anyone looking to perform Monte Carlo simulation techniques on their projects, should seriously consider Full Monte!

You can download a 30-day free trial copy from Barbecana’s web site here http://barbecana.com/free-trial/ and check it out for yourself. Stating the obvious, you will need a copy of Microsoft Project 2007 or 2010!