A simulation model was developed to illustrate the problem with assuming activity times were known with certainty.  However, in the simulation analysis, the activity times were assumed to be normally distributed.  Do you have any concerns with assuming that the activity times were normally distributed?  Redo the simulation analysis using a BetaPERT distribution with parameters of 7, 10, and 15, for the optimistic, most likely, and pessimistic times estimates.  How do the results compare with the results obtained using a normal distribution?