Project Management Stack Exchange is a question and answer site for project managers. It only takes a minute to sign up.
Sign up to join this community
Microsoft Project 2010 (and 2013) no longer contain PERT charts. With this in mind, what is the recommended way to model uncertainty in project time estimates?
UPDATE: I only brought up PERT diagrams because supposedly they were used for representing schedule uncertainty in the past. I don't really care about PERT diagrams. What is the best way to represent schedule uncertainty in Microsoft Project?
The only way to get a true sense of uncertainty in your schedule is to run a Monte Carlo simulation. Seeing a distribution that you created for each work package is meaningless. You need to run hundreds of simulated projects so that the infinite number of combinations of those distributions can play out.
As far as I can tell, there is no out of the box simulator in MS Project. But there are add-ins that you can purchase such as @Risk, Fullmonte, and others.
Microsoft Project 2010 does have PERT charts. It is now called by its generic name "Network Diagram". You can select it from the "View" menu. I don't have MS Project 2013, but I suspect it will be available by the same name "Network Diagram". Checkout whether that is what you are looking for.
I am afraid the Pessimistic, Optimistic and Most Likely estimates have gone away in MS Project 2010. I found that somebody has developed an add-on he claims will help you do this. I have not used this add-on, nor do I know whether it is being maintained: http://blogs.technet.com/b/projectified/archive/2009/11/24/3296207.aspx
I'm skeptical that project can do what you ask; I'm not positive, but I'm skeptical. Let me offer a few options.
First, you could use several of the custom fields to represent minimum, maximum and most likely estimates. (e.g. num1, num2, num3). If I were in your shoes, I'd probably allocate three more fields to indicate the probability, impact and aggregate risk rating for the task, and use the combination of those to develop a dynamic adjustment to schedule. I've worked with math & formulas in project, and I'd strongly suggest that you either purchase the add-ins suggested by @david Espina, or else export to excel for the monte carlo simulation. My preference here is reinforced by your statement that you want to see a graphical representation of the confidence bars; I doubt you can do that in project, but you should be able to do so in excel.
Alternatively, if your real goal is to manage the aleatory risk intrinsic to scheduling, you could adopt Critical chain scheduling, which (at a rough summary) pushes all the risk & slack to a buffer at the end of the project, and consolidates the management.
There is a third option - what you are describing as a PERT chart is not the PERT chart that I'm familiar with. If you check Wikipedia on PERT, they describe the PERT as equivalent to the network diagram referenced by @Ashok Ramachandran, but they do reference using Microsoft Visio to create node diagrams that are related to the PERT estimation techiques you reference. Of course it doesn't lead to the error display bars you're looking for.
I think if you want what you've asked for, you're going to have to cobble it together from multiple products. In my personal opinion, the resulting schedule will be quite difficult to maintain. If you do develop such a product and document the procedures you use to maintain the schedule, I think it would be worth publishing. I can envision value in that kind of a scheduling tool and associated procedures.
I emailed Steve McConnell, author of Rapid Development. He explained that the optimistic/pessimistic multipliers found in his book (table 8-1 "Estimate Multipliers by Project Phase") are meant to be applied on top of the "most likely" single-point estimate. In other words, say you have a total effort of 8 staff months and 6 months schedule, you end up with 2-32 staff months and 3.6-9.6 month schedule. You can apply the same logic to Microsoft Project. In my case I created a macro that multiplied the "Work" of all tasks by a factor of 0.25 or 4.0 and noted the resulting schedule.
