Formalisms for modelling the discrete manufacturing systems

Considering that in a discrete manufacturing system the production rates of machines are stochastic and not equal, we focus on transient analysis of Markov models. The system is decomposed resulting in reduced computation burden into several base-elements, which determine the event occurrence such as: a machine fails, a machine is repaired, a buffer fills up, a buffer empties. The global properties of machines are determined with the computation of distribution of time to absorption in Markov models of manufacturing systems with deadlocks of failures. It is shown that a discrete-event system formulation and state partition into basic cells and fast and slow varying section, lead to a reduced computation cost.