In production systems there are often capacity oriented performance objectives, like a desired total throughput, a desired average throughput time and average work in-process. Such performance objectives are expressed in "units of products" rather than in specific product types. This book presents a way of modeling and analyzing production systems so, that such capacity oriented performance criteria can be measured in a simple way. The model consists of three basic elements. 1. The product types in the system are aggregated. 2. The product flow is modeled as being continuous. 3. The machines in the model have a finite number of states. Each state has a phase-type sojourn distribution and an associated production speed. Transitions between the states are determined by an irreducible Markov transition matrix. In the book both the mathematical properties and the practical applicabilities of the model are investigated. The model is extensively analyzed for various layouts, like flow lines, assembly disassembly systems and networks where parallel machines share common buffers. Furthermore various ways of controlling the product flow in the model are investigated, such as Base Stock Control, Workload Control, control by finite buffers and control by the Reorder Point System. An approximation technique is developed for a quick estimation of performance measures like throughput and average work-in-process, for networks with layouts and control techniques like those above-mentioned.