Civil Engineering Reference
In-Depth Information
The critical path identified in the deterministic approach may not be the same
as those in the probabilistic critical path when risks are incorporated into the
schedule.
The combination of durations, where pessimistic durations may stack along
a path—because of logic relationships—will significantly extend the predicted
completion date.
When discussing duration uncertainty, many project managers believe that their
initial estimations of durations are accurate and that, if the risk discussion continues,
their durations are the “most likely” durations. This seems logical, and so the neces-
sary discussion about the difference between deterministic and probabilistic schedule
development rarely happens. Since the durations are estimates of future events, there
is no certainty that the estimates will be accurate. In addition, duration estimates are
based on predicted crew production rates. Crew production rates are based on a num-
ber of variables, including the composition of the crew, the crew's level of knowledge
about the specific activity, weather conditions, the availability and condition of the
equipment, crowding of the workspace, and many others. All of these factors can affect
the actual duration of the activity, making it unlikely that the original estimate can be
achieved in some cases. An additional problem with using deterministic approaches
with no risk analysis is that many assumptions are made during the early development
of the schedule, but rarely, if ever, are they questioned or reviewed again during the
project. The assumptions themselves can lead to higher risks if they prove to be false
or inadequate.
So, a probabilistic approach takes the concept of estimating durations based on
average production rates to the next level; with this approach, we look at the range of
possible project durations based on a spread (called distribution in statistics) in the esti-
mated activity durations, commonly called a three-point estimate .Theterm three-point
estimate comes from its general definition, which includes three duration estimates:
the pessimistic, the most likely, and the optimistic. This allows for a compiling of all
general risks to the duration into a three-point estimate by providing a range of dura-
tions that can be used in the risk analysis. It also helps with accuracy when there is a
higher level of unknowns about the project, as is often the case in the early stages of any
project but especially of design-build projects, where the design is under development
atthesametimeastheschedule.
There is also a risk of the stacking of inaccurate or changed durations along the
same path, which would magnify the uncertainty of duration estimating. With the
correct risk management approach, the uncertainty of activity durations is more likely
to be taken into account in the schedule's predictions.
The use of the Monte Carlo simulations makes general duration uncertainties rela-
tively easy to analyze, as well as risks from activity relationships and some specific event
risks. Monte Carlo analysis runs a large number of iterations based on the spread of the
duration estimates so that many combinations of durations are used. This probabilistic
approach recognizes that the more appropriate way to model durations is through the
use of statistics, by which if enough iterations are run, the results will generally fall
Search WWH ::




Custom Search