Guild of Project Controls: Compendium | Roles | Assessment | Certifications | Membership

Tips on using this forum..

(1) Explain your problem, don't simply post "This isn't working". What were you doing when you faced the problem? What have you tried to resolve - did you look for a solution using "Search" ? Has it happened just once or several times?

(2) It's also good to get feedback when a solution is found, return to the original post to explain how it was resolved so that more people can also use the results.

Duration Correlation of activities in a Risk Sched

2 replies [Last post]
Colin Cropley
User offline. Last seen 2 years 46 weeks ago. Offline
Joined: 1 Feb 2002
Posts: 57
Groups: None
I wish to ask forum contributors to address the question "How do you decide which activities in a risk schedule are in reality duration-correlated and by what percentage?"
This is a key question for obtaining reliable results from Schedule Risk Analysis (SRA). Without correlation, as you divide activities into smaller parts, provided you keep the proportions of uncertainty on the activity durations, SRA analysis will show the schedule duration risk to be smaller and smaller. However, from practical experience we know this is not true and the reason is correlation.

SRA will assume each of those sub-divided activities is completely independent of the other activities, whereas they are +100% duration correlated, if you have simply divided an activity into smaller pieces in series.

There is a related question that I will ask separately.


Philip Rawlings
User offline. Last seen 5 years 11 weeks ago. Offline
Joined: 16 Nov 2004
Posts: 41
Groups: None
No simple answer I’m afraid. There has to be some subjectivity. I usually start by running with correlation at 100% and at 0% (easy to do in pertmaster with a simple macro) - this gives an appreciation of the problem, whether it’s worth investment of time or not. Then I correlate similar sets of tasks (engineering, procurement etc), then I correlate across project for global factors (PM is the biggest driver).

Last (and most important) I see whether the results makes sense, then I show it to the stakeholders to ensure it is believable.
activities that use the same resources have correlated durations. In our software Spider Project duration is calculated by dividing activity volume by resource productivity. Duration uncertainty is usually defined by resource productivity uncertainty.
This correlation is not single but most obvious.
Using Monte Carlo simulation you shall choose what to simulate though it will not work properly for large projects in any case.