Service Systems are nowadays considered as the most profitable assets of both Private Companies and Public Organizations, since they offer a highly added value for many different kind of products, tangible and intangible. Among them, a Call-Center is considered to be as very difficult both to understand and hence to perfectly manage. This is mainly due to its intrinsic dynamics and to the unpredictable behaviour of input data. Furthermore, modern Call-Centers tend to focus on the management of various different types of calls, which usually require the government of different situations together with a different agents' skill management. In such a scenario, the main objective is to find an efficient way to organize and manage a multi-task Call-Center. We will show how these kind of Call-Centers can efficiently be managed through a Dynamic Skill-Based Routing policy, instead of a Static Skill-Based Routing or even a call-processing policy using separate trunks-lines. After describing the typical structures of Call-Centers, we’ll define a SD model, stating again how a simulation model can be useful in the management and sizing of a Call-Center. The paper will end with a case study where, through the simulation of different policies, it will be shown how the Dynamic Skill-Based Routing is the most efficient way to manage multi-call-types Call-Centers