Describe the scenarios when it would be preferable to use simulation instead of analytical models, and scenarios where it would be preferable to use analytical models instead of simulation. Give examples wherever appropriate.

The question on whether to use analytical or simulation approaches has stemmed from the recent advances in computation. Before the advent of powerful computers, analytical approaches were often the only way to solve problems. But recent advances in computation have granted often “easier” ways to solve complex mathematical models. Of course, both approaches stem from mathematical models (as opposed to physical models) and the hierarchical structure of different ways to study a system is given in Figure 1. Figure 1: Ways to Study a System [4]

The term easier is a rather loosely defined one. Easier is meant by the method of compiling information, where highly complex systems can now be simulated by advanced computing software. Simulation can have many pitfalls which will be explained later, and is by no means an easy process. Basically, once a mathematical model is built to describe a system, it must be tested to see how it can be used to answer various questions of interest from the system it represents [4]. If this mathematical representation is simple enough, it is possible to work with the underlying quantities and

*…show more content…*

But once it is decided that the system at hand is too complex to derive an analytical solution directly, simulation must be the only choice. There are advantages to simulation, disadvantages, and often pitfalls. The first advantage has been already discussed and is the ability to study a system that was too complex to derive an analytical solution. Other advantages of simulation also allow time compression, maintenance of experimental conditions, and alternative system configuration comparison