This is a difficult question to answer as you haven’t provided any details about the nature of the tasks. In this circumstance, normally this would not be considered enough data with which to construct a reliable distribution. Most likely with such a limited data set, you will have a good indicators for what the boundary conditions should be (Low and High) or where your Most Likely lies. In this case, I would lean more heavily on your SME’s experience as to setting a distribution.
Intaver Support Team
Intaver Institute Inc.
Home of Project Risk Management and Project Risk Analysis software RiskyProjectwww.intaver.com