Different uncertainties
Uncertainties are different in nature, so they have to be handled and described differently. There is a rich set of tools available for model builders to achieve this. The differences have to be handled with substantial experience, otherwise even apparently correct calculations can lead to false results. In essence, such differences arise from the differences of the applicable statistical methods.
Thus, the methods consist in the various procedures of statistics and probability theory. The sensitivity of the applications generally requires that the system engineer consult with the experts of the relevant field even in the case of simpler problems.
Fuzzy methods are popular procedures for handling uncertainties, though their use is a controversial issue partly because they are handled similarly to probability calculus methods and partly due to the uncertainties relating to definitions. They are particularly significant because of their relation to natural language processing.