Slide 1: Probability and Activity Times
Slide 1: Probability and Activity Times
Let's begin with single activity time estimates. In our example, building a house, we assume
activity times are known with certainty, meaning we have a single estimate with no significant
variation. For instance, laying the foundation is estimated to take 2 months, assuming no
deviation from this time."
"In reality, it's rare to have such certainty in activity time estimates. Most project activities are
unique and lack sufficient historical data for accurate prediction. This brings us to a comparison
between CPM and PERT. The Critical Path Method, or CPM, uses deterministic times, whereas
the Program Evaluation and Review Technique, or PERT, uses probabilistic activity times."
"Today's discussion will focus on using probabilistic methods for estimating activity times in
project networks. We'll explore how these methods can provide more realistic and flexible time
estimates compared to single deterministic estimates."
"Next, let's look at project crashing. For example, if an original activity duration is 12 weeks at a
cost of $3000, and we want to crash it to 7 weeks at a cost of $5000, the total crash cost would
be $2000. This is calculated by subtracting the normal cost from the crash cost. The crash time
is 5 weeks, and the crash cost per week is $400, which is the total crash cost divided by the
crash time."
"Now, we turn to the PERT methodology. Traditionally, a single time estimate is assigned to
each activity, but PERT uses three estimates to model uncertainty: the most likely time, the
optimistic time, and the pessimistic time. These estimates help calculate the mean and variance
for a beta distribution of activity times."
"The beta distribution is chosen for its flexibility and ability to approximate mean and variance
using three estimates. It’s a continuous distribution without a predetermined shape, meaning it
can be skewed based on the estimates provided. This makes it a traditional choice for
probabilistic network analysis."
"Time estimates in PERT are subjective. The most likely time is the most frequent occurrence if
the activity were repeated many times. The optimistic time is the shortest possible time under
ideal conditions, and the pessimistic time is the longest possible time under the worst
conditions."
"Using these three time estimates, we can calculate the mean and variance of a beta
distribution. Here, 'a' represents the optimistic time estimate, 'm' the most likely time estimate,
and 'b' the pessimistic time estimate."
Slide 10 and 11 :
Using the normal distribution, probabilities can be determined by computing the number
of standard deviations
As an example of the computation of the individual activity mean times and variances,
consider activity 1. The three time estimates 1a = 6, m = 8, b = 102 are substituted in our
beta distribution formulas and it follows
Slide : Summary
"In summary, using probabilistic time estimates such as those in PERT allows for a more flexible
and realistic approach to project management. This helps in planning and managing project
timelines more effectively by accounting for uncertainties and variations in activity times."