From Compsci.ca Wiki
Simulation is a method to perform experiments which may not be possible or feasible in the real world.
Benefits of Simulation
The reasons why the real world experiment may not be performed may vary. For example, an experiment may otherwise be too costly (in terms of equipment), too dangerous (disaster scenarios, extreme environments), or to save time. A simulation can often run much faster than an experiment would in real life. This is helpful in many cases such as environmental simulation since the amount of time being modelled is often very large. In applied computing, such as wireless network design, simulation can help to determine which designs are most promising before implementation in real equipment which often saves time, effort and money.
Drawbacks of Simulation
At the same time, simulation is often criticized for not being a realistic representation of the real world. In wireless network simulation for example, critics often argue that performance in the real world is not nearly as good as the simulations often predict. However, some of these problems may be eventually improved with more realistic models. Another drawback of simulation is reproducibility of experiments. Unless there are standardized simulation tools being used, it is often difficult to compare or evaluate results. At the same time, the standardized tools often do not provide the functionality required for a given experiment so customization is necessary.