Decomposition in analytic modeling is an extension of the analytical approach. We separate them because the parameters and functions are used here to characterize pieces of the parallel system. The idea behind decomposition is to simplify the modeling process. It also allows modular models, where the various pieces can be updated and replaced independently from the others.
Parallel systems can be decomposed in various ways for modeling purposes. Next we present some of the options, and their advantages and disadvantages.