Some of the risk
analysis techniques:
- Brainstorming: this can be used in
formative project planning & can be used to identify &
postulate risk scenarios for a particular project. The advantage of
this method is that it is very simple & effective & the
weakness here is that each member must build ideas on the preceding
comments.
- Sensitivity analysis: this technique
seeks to place value on the effect of change of a single variable
within the project by analyzing the effect on the project plan.
This is considered to be simplest form of risk analysis. Other
benefits are realistic decision making, impressing management,
range possible outcomes. Weaknesses are that variables are treated
individually & the sensitivity diagram would give no indication
of probability of occurrence.
- Probability analysis: this specifies
the probability distribution of each variable & then considers
situations where any or all of these variables can be changed at
the same time. The befit here is that this method overcomes the
method of sensitivity analysis. The weakness of this method is
Defining the probability of occurrence of any specific variable may
be quite difficult, particularly as political or commercial
environments can change quite rapidly.
- Delphi method: The basic concept is
to derive a consensus using a panel of experts to arrive at a
convergent solution to a specific problem. This is particularly
useful in arriving at probability assessments relating to future
events where the risk impacts are large and critical. The first and
vital step is to select a panel of individuals who have experience
in the area at issue. For best results, the panel members should
not know each other identity and the process should be conducted
with each at separate locations. The responses, together with
opinions and justifications, are evaluated and statistical feedback
is furnished to each panel member in the next iteration. The
process is continued until group responses converge to s specific
solution
- Decision tree analysis: in this case
there would be number of opinions that would be available in the
course of reaching the final results. The strength of this method
is that it considers the probability of each outcome. The
likelihood of failure is quantified & value is given to each
decision. The weakness is that this decision is usually applied to
cost & time considerations.
- Monte Carlo: The Monte Carlo method,
simulation by means of random numbers, provides a powerful yet
simple method of incorporating probabilistic data. Basic steps
are:
a. Assess the range of the
variables being considered and determine the probability
distribution most suited to each.
b. For each variable within its
specific range, select a value randomly chosen, taking account of
the probability distribution for the occurrence of the
variable.
c. Run a deterministic
analysis using the combination of values selected for each one of
the variables.
d. Repeat steps 2 and 3 a number
of times to obtain the probability distribution of the result.
Typically between 100 and 1000 iterations are required depending on
the number of variables and the degree of confidence required.