New technological trends lead to the increasing use of network technologies in automation. Especially the Ethernet with TCP/IP and wireless networks find growing acceptance. The resulting networked automation systems (NAS) display properties such as stochastic delays and information loss, which are not known in classical automation structures. When control quality is to be assessed, these properties have to be determined. In this paper, the determination is achieved in a modeling approach based on Probabilistic Timed Automata (PTA). The derived models allow the analysis of delays using Probabilistic Model Checking (PMC). A case study will illustrate how the results of the analysis can be applied to increase the product quality in a manufacturing system controlled by an NAS.