COMPARISONOFMETHODS

Monte Carlo simulation has more intuitive appeal than does the generation of system

moments and consequentlyis easier tounderstand.

The desired precision can be

obtained by conducting sufficient trials.

Also, theMonte Carlo method is veryflexible

and can be applied to many highly complex situations for which the method of generation of

system moments becomes too difficult.

This is especially true when there are

interrelationships between thecomponent variables.

A major drawback of the Monte Carlomethod isthat there isfrequentlyno way of determining whether any of the variables aredominant or more important than

others. redone.

Furthermore, if achange is made in one variable, theentire simulation must be Also,the method generally requires developing acomplex computer

program; andif a large number of trials are required, agreat deal of computer timemay be neededto obtain the necessary answers.

Consequently, the generation of system moments, in conjunction with a Pearson or Johnson distribution approximation, issometimesthe most economical approach. Although the precision of the answers usually cannot be easily assessed for this method,

the results of the study

suggest that this approach oftendoes provide an adequate

approximation.

In addition, thegeneration of system moments allowsus to analyzethe

relative importance of each component variableby examining the magnitude of its partial derivative.